Merge pull request #4423 from joycech333/feat/inception-llm-service

feat: add Inception LLM service with Mercury 2 support

.agents/skills/changelog

@@ -0,0 +1 @@
+../../.claude/skills/changelog

.agents/skills/cleanup

@@ -0,0 +1 @@
+../../.claude/skills/cleanup

.agents/skills/code-review

@@ -0,0 +1 @@
+../../.claude/skills/code-review

.agents/skills/docstring

@@ -0,0 +1 @@
+../../.claude/skills/docstring

.agents/skills/pr-description

@@ -0,0 +1 @@
+../../.claude/skills/pr-description

.agents/skills/pr-submit

@@ -0,0 +1 @@
+../../.claude/skills/pr-submit

.agents/skills/update-docs

@@ -0,0 +1 @@
+../../.claude/skills/update-docs

.claude/skills/cleanup/SKILL.md

@@ -1,3 +1,8 @@
+---
+name: cleanup
+description: Review, refactor, document, and validate code changes in the current branch
+---
+
 # Code Cleanup Skill
 
 The **Code Cleanup Skill** reviews, refactors, and documents code changes in your current branch, ensuring alignment with **Pipecat's architecture, coding standards, and example patterns**.

.claude/skills/squash-commits/SKILL.md

@@ -0,0 +1,91 @@
+---
+name: squash-commits
+description: Reorganize messy branch commits into a small set of logical, meaningful commits without changing any content. Drops merge-from-main commits. Safe: creates a backup branch first.
+---
+
+Reorganize the commits on the current branch into a small number of logical commits. Do NOT change any file content — only the commit structure changes.
+
+## Instructions
+
+### 1. Safety check
+
+```bash
+git status --short
+```
+
+If there are uncommitted changes, stop and tell the user to commit or stash them first.
+
+### 2. Inspect the branch
+
+```bash
+git log main..HEAD --oneline
+git diff main..HEAD --name-only
+```
+
+List every file changed vs `main` and every commit on the branch (excluding merge commits from main).
+
+### 3. Create a backup branch
+
+```bash
+git branch backup/<current-branch-name>
+```
+
+Tell the user the backup exists so they can recover if needed.
+
+### 4. Soft-reset to main and unstage everything
+
+```bash
+git reset --soft main
+git restore --staged .
+```
+
+All branch changes are now in the working tree, unstaged. No content has changed.
+
+### 5. Plan the logical groups
+
+Read the changed files and the original commit messages to understand what the work covers. Group related files into logical commits. Typical groups:
+
+- Core feature or fix (new source files + modified core files)
+- Secondary features or fixes (each as its own commit if distinct)
+- Refactoring or renames
+- Tests
+- Changelogs / docs
+
+Use the changelog files (if any) as a strong hint — each changelog entry often maps to one commit.
+
+Present the proposed grouping to the user and ask for confirmation before committing.
+
+### 6. Commit in logical groups
+
+For each group, stage only the relevant files and commit with a clear message following the project's conventions:
+
+```bash
+git add <file1> <file2> ...
+git commit -m "..."
+```
+
+Use conventional commit prefixes if the project uses them (`feat:`, `fix:`, `refactor:`, `test:`, `chore:`).
+
+### 7. Verify
+
+```bash
+git log main..HEAD --oneline
+git diff main..HEAD --name-only
+git status --short
+```
+
+Confirm:
+- Commit count is small and each message is meaningful
+- The set of changed files vs `main` is identical to before
+- Working tree is clean
+
+### 8. Remind about force-push
+
+The branch history has been rewritten. Tell the user they will need to `git push --force-with-lease` when they are ready to update the remote. Do NOT push automatically.
+
+## Rules
+
+- Never change file contents. If you find yourself editing a file, stop.
+- Never skip the backup branch step.
+- Never force-push without explicit user instruction.
+- If any step fails or the result looks wrong, tell the user and suggest restoring from the backup: `git reset --hard backup/<branch-name>`.

.github/workflows/coverage.yaml

@@ -42,6 +42,7 @@ jobs:
             --extra langchain \
             --extra livekit \
             --extra piper \
+            --extra runner \
             --extra sagemaker \
             --extra tracing \
             --extra websocket

.github/workflows/format.yaml

@@ -32,7 +32,9 @@ jobs:
         run: uv python install 3.12
 
       - name: Install development dependencies
-        run: uv sync --group dev --extra daily --extra tracing
+        # `--all-extras` (matching the dev setup in README.md) so pyright can
+        # resolve types from various optional dependencies.
+        run: uv sync --group dev --all-extras --no-extra gstreamer --no-extra local
 
       - name: Ruff formatter
         id: ruff-format

.github/workflows/tests.yaml

@@ -46,6 +46,7 @@ jobs:
             --extra langchain \
             --extra livekit \
             --extra piper \
+            --extra runner \
             --extra sagemaker \
             --extra tracing \
             --extra websocket

AGENTS.md

@@ -0,0 +1,174 @@
+# AGENTS.md
+
+This file provides guidance to AI coding agents when working with code in this repository.
+
+## Project Overview
+
+Pipecat is an open-source Python framework for building real-time voice and multimodal conversational AI agents. It orchestrates audio/video, AI services, transports, and conversation pipelines using a frame-based architecture.
+
+## Common Commands
+
+```bash
+# Setup development environment
+uv sync --group dev --all-extras --no-extra gstreamer --no-extra local
+
+# Install pre-commit hooks
+uv run pre-commit install
+
+# Run all tests
+uv run pytest
+
+# Run a single test file
+uv run pytest tests/test_name.py
+
+# Run a specific test
+uv run pytest tests/test_name.py::test_function_name
+
+# Preview changelog
+uv run towncrier build --draft --version Unreleased
+
+# Lint and format check
+uv run ruff check
+uv run ruff format --check
+
+# Update dependencies (after editing pyproject.toml)
+uv lock && uv sync
+```
+
+## Architecture
+
+### Frame-Based Pipeline Processing
+
+All data flows as **Frame** objects through a pipeline of **FrameProcessors**:
+
+```
+[Processor1] → [Processor2] → ... → [ProcessorN]
+```
+
+**Key components:**
+
+- **Frames** (`src/pipecat/frames/frames.py`): Data units (audio, text, video) and control signals. Flow DOWNSTREAM (input→output) or UPSTREAM (acknowledgments/errors).
+
+- **FrameProcessor** (`src/pipecat/processors/frame_processor.py`): Base processing unit. Each processor receives frames, processes them, and pushes results downstream.
+
+- **Pipeline** (`src/pipecat/pipeline/pipeline.py`): Chains processors together.
+
+- **ParallelPipeline** (`src/pipecat/pipeline/parallel_pipeline.py`): Runs multiple pipelines in parallel.
+
+- **Transports** (`src/pipecat/transports/`): Transports are frame processors used for external I/O layer (Daily WebRTC, LiveKit WebRTC, WebSocket, Local). Abstract interface via `BaseTransport`, `BaseInputTransport` and `BaseOutputTransport`.
+
+- **Pipeline Task (`src/pipecat/pipeline/task.py`)**: Runs and manages a pipeline. Pipeline tasks send the first frame, `StartFrame`, to the pipeline in order for processors to know they can start processing and pushing frames. Pipeline tasks internally create a pipeline with two additional processors, a source processor before the user-defined pipeline and a sink processor at the end. Those are used for multiple things: error handling, pipeline task level events, heartbeat monitoring, etc.
+
+- **Pipeline Runner (`src/pipecat/pipeline/runner.py`)**: High-level entry point for executing pipeline tasks. Handles signal management (SIGINT/SIGTERM) for graceful shutdown and optional garbage collection. Run a single pipeline task with `await runner.run(task)` or multiple concurrently with `await asyncio.gather(runner.run(task1), runner.run(task2))`.
+
+- **Services** (`src/pipecat/services/`): 60+ AI provider integrations (STT, TTS, LLM, etc.). Extend base classes: `AIService`, `LLMService`, `STTService`, `TTSService`, `VisionService`.
+
+- **Serializers** (`src/pipecat/serializers/`): Convert frames to/from wire formats for WebSocket transports. `FrameSerializer` base class defines `serialize()` and `deserialize()`. Telephony serializers (Twilio, Plivo, Vonage, Telnyx, Exotel, Genesys) handle provider-specific protocols and audio encoding (e.g., μ-law).
+
+- **RTVI** (`src/pipecat/processors/frameworks/rtvi.py`): Real-Time Voice Interface protocol bridging clients and the pipeline. `RTVIProcessor` handles incoming client messages (text input, audio, function call results). `RTVIObserver` converts pipeline frames to outgoing messages: user/bot speaking events, transcriptions, LLM/TTS lifecycle, function calls, metrics, and audio levels.
+
+- **Observers** (`src/pipecat/observers/`): Monitor frame flow without modifying the pipeline. Passed to `PipelineTask` via the `observers` parameter. Implement `on_process_frame()` and `on_push_frame()` callbacks.
+
+### Important Patterns
+
+- **Context Aggregation**: `LLMContext` accumulates messages for LLM calls; `UserResponse` aggregates user input
+
+- **Turn Management**: Turn management is done through `LLMUserAggregator` and
+  `LLMAssistantAggregator`, created with `LLMContextAggregatorPair`
+
+- **User turn strategies**: Detection of when the user starts and stops speaking is done via user turn start/stop strategies. They push `UserStartedSpeakingFrame` and `UserStoppedSpeakingFrame` respectively.
+
+- **Interruptions**: Interruptions are usually triggered by a user turn start strategy (e.g. `VADUserTurnStartStrategy`) but they can be triggered by other processors as well, in which case the user turn start strategies don't need to. An `InterruptionFrame` carries an optional `asyncio.Event` that is set when the frame reaches the pipeline sink. If a processor stops an `InterruptionFrame` from propagating downstream (i.e., doesn't push it), it **must** call `frame.complete()` to avoid stalling `push_interruption_task_frame_and_wait()` callers.
+
+- **Uninterruptible Frames**: These are frames that will not be removed from internal queues even if there's an interruption. For example, `EndFrame` and `StopFrame`.
+
+- **Events**: Most classes in Pipecat have `BaseObject` as the very base class. `BaseObject` has support for events. Events can run in the background in an async task (default) or synchronously (`sync=True`) if we want immediate action. Synchronous event handlers need to execute fast.
+
+- **Async Task Management**: Always use `self.create_task(coroutine, name)` instead of raw `asyncio.create_task()`. The `TaskManager` automatically tracks tasks and cleans them up on processor shutdown. Use `await self.cancel_task(task, timeout)` for cancellation.
+
+- **Error Handling**: Use `await self.push_error(msg, exception, fatal)` to push errors upstream. Services should use `fatal=False` (the default) so application code can handle errors and take action (e.g. switch to another service).
+
+### Key Directories
+
+| Directory                  | Purpose                                            |
+| -------------------------- | -------------------------------------------------- |
+| `src/pipecat/frames/`      | Frame definitions (100+ types)                     |
+| `src/pipecat/processors/`  | FrameProcessor base + aggregators, filters, audio  |
+| `src/pipecat/pipeline/`    | Pipeline orchestration                             |
+| `src/pipecat/services/`    | AI service integrations (60+ providers)            |
+| `src/pipecat/transports/`  | Transport layer (Daily, LiveKit, WebSocket, Local) |
+| `src/pipecat/serializers/` | Frame serialization for WebSocket protocols        |
+| `src/pipecat/observers/`   | Pipeline observers for monitoring frame flow       |
+| `src/pipecat/audio/`       | VAD, filters, mixers, turn detection, DTMF         |
+| `src/pipecat/turns/`       | User turn management                               |
+
+## Code Style
+
+- **Docstrings**: Google-style. Classes describe purpose; `__init__` has `Args:` section; dataclasses use `Parameters:` section.
+- **Deprecations**: Use the `.. deprecated:: <version>` Sphinx directive in docstrings (never inline tags like `[DEPRECATED]`), and pair it with a runtime `warnings.warn(..., DeprecationWarning)` at the call site. See `CONTRIBUTING.md` for full conventions.
+- **Linting**: Ruff (line length 100). Pre-commit hooks enforce formatting.
+- **Type hints**: Required for complex async code.
+- **Dataclass vs Pydantic**: Use `@dataclass` for frames and internal pipeline data (high-frequency, no validation needed). Use Pydantic `BaseModel` for configuration, parameters, metrics, and external API data (benefits from validation and serialization). Specifically:
+  - `@dataclass`: Frame types, context aggregator pairs, internal data containers
+  - `BaseModel`: Service `InputParams`, transport/VAD/turn params, metrics data, API request/response models, serializer params
+
+### Docstring Example
+
+```python
+class MyService(LLMService):
+    """Description of what the service does.
+
+    More detailed description.
+
+    Event handlers available:
+
+    - on_connected: Called when we are connected
+
+    Example::
+
+        @service.event_handler("on_connected")
+        async def on_connected(service, frame):
+            ...
+    """
+
+    def __init__(self, param1: str, **kwargs):
+        """Initialize the service.
+
+        Args:
+            param1: Description of param1.
+            **kwargs: Additional arguments passed to parent.
+        """
+        super().__init__(**kwargs)
+
+
+# Pydantic params class with a deprecated field
+class MyParams(BaseModel):
+    """Configuration parameters for MyService.
+
+    Parameters:
+        new_setting: Replacement for ``old_setting``.
+        old_setting: Legacy setting, no longer used.
+
+            .. deprecated:: 1.2.0
+                Use ``new_setting`` instead. Will be removed in 2.0.0.
+    """
+
+    new_setting: str = "default"
+    old_setting: str | None = None
+```
+
+## Service Implementation
+
+When adding a new service:
+
+1. Extend the appropriate base class (`STTService`, `TTSService`, `LLMService`, etc.)
+2. Implement required abstract methods
+3. Handle necessary frames
+4. By default, all frames should be pushed in the direction they came
+5. Push `ErrorFrame` on failures
+6. Add metrics tracking via `MetricsData` if relevant
+7. Follow the pattern of existing services in `src/pipecat/services/`
+
+## Testing
+
+Test utilities live in `src/pipecat/tests/utils.py`. Use `run_test()` to send frames through a pipeline and assert expected output frames in each direction. Use `SleepFrame(sleep=N)` to add delays between frames.

CHANGELOG.md

@@ -7,6 +7,515 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 <!-- towncrier release notes start -->
 
+## [1.2.1] - 2026-05-15
+
+### Changed
+
+- Changed the default WebSocket endpoints for `GradiumSTTService` and
+  `GradiumTTSService` to the region-neutral
+  `wss://api.gradium.ai/api/speech/asr` and
+  `wss://api.gradium.ai/api/speech/tts`. Gradium now automatically routes
+  traffic to the nearest endpoint. Override the url to pin to a specific
+  region.
+  (PR [#4500](https://github.com/pipecat-ai/pipecat/pull/4500))
+
+### Fixed
+
+- Fixed bot hangs when `filter_incomplete_user_turns` was enabled and the LLM
+  responded by calling a tool. The user turn never finalized, so the assistant
+  aggregator gated the tool-result context push and the LLM continuation never
+  ran. Tool calls now finalize the turn the moment they start, before the
+  function dispatches.
+  (PR [#4501](https://github.com/pipecat-ai/pipecat/pull/4501))
+
+## [1.2.0] - 2026-05-14
+
+### Added
+
+- Added a `session_id` field to `RunnerArguments` so bots can log or trace a
+  per-session identifier in local development the same way they can in Pipecat
+  Cloud. The development runner now mints a UUID at every construction site,
+  and paths that already returned a `sessionId` to the caller (Daily `/start`,
+  dial-in webhook) share that same UUID with the runner args instead of
+  generating two. The SmallWebRTC `/api/offer` endpoint also accepts an
+  optional `session_id` query parameter so the `/sessions/{session_id}/...`
+  proxy can thread it through.
+  (PR [#4385](https://github.com/pipecat-ai/pipecat/pull/4385))
+
+- Added a `max_buffer_delay_ms` constructor argument to `CartesiaTTSService`
+  for controlling Cartesia's server-side text buffering. When unset, Pipecat
+  picks a sensible default based on `text_aggregation_mode`: `0` in `SENTENCE`
+  mode (custom buffering — avoids stacking client-side aggregation on top of
+  Cartesia's default 3000ms server buffer) and unset in `TOKEN` mode
+  (Cartesia's managed buffering applies). Pass an explicit value (0–5000ms) to
+  override.
+  (PR [#4390](https://github.com/pipecat-ai/pipecat/pull/4390))
+
+- Added a `mip_opt_out` constructor argument to `DeepgramTTSService` and
+  `DeepgramHttpTTSService` so callers can opt out of the Deepgram Model
+  Improvement Program. When set, the value is forwarded to Deepgram as a query
+  parameter on the speak request. Defaults to `None`, which preserves the
+  existing behavior. See https://dpgr.am/deepgram-mip for pricing implications
+  before enabling.
+  (PR [#4400](https://github.com/pipecat-ai/pipecat/pull/4400))
+
+- Added an opt-in `add_tool_change_messages` flag to the LLM aggregators (set
+  via `LLMContextAggregatorPair(..., add_tool_change_messages=True)`) that
+  appends a developer-role message to the context whenever `LLMSetToolsFrame`
+  changes the set of advertised standard tools. Helps the LLM stay coherent
+  across mid-conversation tool changes, mitigating several flavors of
+  tool-call-related hallucination: calling tools that have been removed,
+  avoiding tools that have been re-added, and hallucinating output (made-up
+  answers or tool-call-shaped non-tool-calls) when tools are unavailable.
+  (PR [#4404](https://github.com/pipecat-ai/pipecat/pull/4404))
+
+- Added `deferred(strategy)` and `DeferredUserTurnStopStrategy` in
+  `pipecat.turns.user_stop`. Wraps a stop strategy so it fires only the
+  inference-triggered event and suppresses `on_user_turn_stopped`, leaving
+  finalization to another strategy in the chain such as
+  `LLMTurnCompletionUserTurnStopStrategy`.
+  (PR [#4405](https://github.com/pipecat-ai/pipecat/pull/4405))
+
+- Added `ExternalUserTurnCompletionStopStrategy` in `pipecat.turns.user_stop` —
+  a generic stop strategy that finalizes the user turn whenever a
+  `UserTurnInferenceCompletedFrame` arrives, regardless of which component
+  produced it. `LLMTurnCompletionUserTurnStopStrategy` now extends this base;
+  future producers (Flux, custom end-of-turn classifiers, etc.) can use the
+  base directly or subclass it to add producer-specific setup.
+  (PR [#4405](https://github.com/pipecat-ai/pipecat/pull/4405))
+
+- Added `on_user_turn_inference_triggered`, a new event on the user turn
+  controller, processor, aggregator and stop strategies that fires when a
+  strategy has enough signal to start LLM inference. By default it fires
+  together with `on_user_turn_stopped`; a gating strategy can fire only the
+  inference-triggered event and defer finalization to a peer.
+  (PR [#4405](https://github.com/pipecat-ai/pipecat/pull/4405))
+
+- Added `FilterIncompleteUserTurnStrategies` in
+  `pipecat.turns.user_turn_strategies` — a `UserTurnStrategies` specialization
+  that wraps the detector chain with `deferred(...)` and appends
+  `LLMTurnCompletionUserTurnStopStrategy` as the finalizer. Common case:
+  `user_turn_strategies=FilterIncompleteUserTurnStrategies()`. Pass
+  `config=UserTurnCompletionConfig(...)` to customize timeouts and prompts.
+  (PR [#4405](https://github.com/pipecat-ai/pipecat/pull/4405))
+
+- Added `LLMTurnCompletionUserTurnStopStrategy` in `pipecat.turns.user_stop`.
+  When installed, the strategy gates `on_user_turn_stopped` on a
+  `UserTurnInferenceCompletedFrame` (a new fieldless system frame emitted by
+  any component that can judge turn completeness — e.g. the
+  `UserTurnCompletionLLMServiceMixin` on `✓`). A `finalization_timeout`
+  provides a safety net if no completion frame ever arrives.
+  (PR [#4405](https://github.com/pipecat-ai/pipecat/pull/4405))
+
+- Added first-class RTVI support for the UI Agent Protocol:
+    - Adds `ui-event`, `ui-snapshot`, and `ui-cancel-task` client-to-server
+  messages, plus `ui-command` and `ui-task` server-to-client messages, with
+  paired `*Data` / `*Message` pydantic models.
+    - Adds built-in command payload models for `Toast`, `Navigate`, `ScrollTo`,
+  `Highlight`, `Focus`, `Click`, `SetInputValue`, and `SelectText`; matching
+  default handlers live in `@pipecat-ai/client-react`.
+    - Adds `RTVIProcessor.on_ui_message` for inbound `ui-event`, `ui-snapshot`,
+  and `ui-cancel-task` messages.
+    - Adds five UI pipeline frames, mirroring the `client-message`
+  frame-and-event pattern: downstream code pushes `RTVIUICommandFrame` /
+  `RTVIUITaskFrame` for the observer to wrap into outbound `UICommandMessage` /
+  `UITaskMessage` envelopes, while the processor pushes inbound
+  `RTVIUIEventFrame`, `RTVIUISnapshotFrame`, and `RTVIUICancelTaskFrame`
+  alongside `on_ui_message`.
+    - Bumps the RTVI `PROTOCOL_VERSION` from `1.2.0` to `1.3.0`.
+  (PR [#4407](https://github.com/pipecat-ai/pipecat/pull/4407))
+
+- AWS Transcribe STT, Polly TTS, Bedrock LLM, and the Bedrock AgentCore
+  processor now resolve credentials via the standard boto3 provider chain (EC2
+  instance profiles, EKS pod roles / IRSA, ECS task roles, SSO,
+  `~/.aws/credentials`) when explicit credentials and `AWS_*` environment
+  variables are absent. Services running with IAM roles no longer need to
+  export static credentials.
+  (PR [#4416](https://github.com/pipecat-ai/pipecat/pull/4416))
+
+- Added `keyterms` support to ElevenLabs STT services so Scribe V2 callers can
+  bias transcription for both file-based and realtime transcription.
+  (PR [#4426](https://github.com/pipecat-ai/pipecat/pull/4426))
+
+- Added `watchdog_min_timeout` parameter to `DeepgramFluxSTT` and
+  `DeepgramFluxSageMakerSTT` (default `0.5` seconds) to control the minimum
+  silence duration before the watchdog sends a silence packet to prevent
+  dangling turns. The actual threshold is `max(chunk_duration * 2,
+  watchdog_min_timeout)`, so it also adapts automatically to the audio chunk
+  size in use.
+  (PR [#4430](https://github.com/pipecat-ai/pipecat/pull/4430))
+
+- Added `cancel_on_interruption=False` support for `GeminiLiveLLMService` on
+  models that support Gemini's NON_BLOCKING tool mechanism (currently Gemini
+  2.x); the conversation now continues while the tool runs. On models that
+  don't yet support NON_BLOCKING (Gemini 3.x), the service surfaces a one-time
+  warning explaining the limitation. (Note: an intermittent 1008 error can
+  occasionally fire on Gemini 2.5 during long-running tool calls; we
+  auto-reconnect.)
+  (PR [#4448](https://github.com/pipecat-ai/pipecat/pull/4448))
+
+- Added `NvidiaSageMakerWebsocketSTTService` for streaming speech recognition
+  using NVIDIA Nemotron ASR via an AWS SageMaker bidirectional-stream endpoint.
+  Produces `InterimTranscriptionFrame` and `TranscriptionFrame` frames, is
+  VAD-aware, and automatically reconnects on error.
+  (PR [#4464](https://github.com/pipecat-ai/pipecat/pull/4464))
+
+- Added NVIDIA Magpie TTS services via AWS SageMaker:
+  `NvidiaSageMakerHTTPTTSService` (single HTTP invocation, streams raw PCM
+  back) and `NvidiaSageMakerWebsocketTTSService` (persistent HTTP/2 bidi-stream
+  with full interruption support via `InterruptibleTTSService`).
+  (PR [#4464](https://github.com/pipecat-ai/pipecat/pull/4464))
+
+- Added support for `reasoning` configuration on `OpenAIRealtimeLLMService`,
+  for use with reasoning-capable Realtime models such as `gpt-realtime-2`.
+  (PR [#4470](https://github.com/pipecat-ai/pipecat/pull/4470))
+
+- Inworld TTS updates:
+    - Added `delivery_mode` setting (`STABLE`/`BALANCED`/`CREATIVE`) to
+  `InworldTTSService` and `InworldHttpTTSService`, enabling the
+  stability-vs-creativity tradeoff in `inworld-tts-2`.
+    - Added language support to `InworldTTSService` and
+  `InworldHttpTTSService`. The `language` setting is now forwarded to the API,
+  and a new `language_to_inworld_language()` helper normalizes Pipecat
+  `Language` enums to Inworld's BCP-47 locale tags.
+  (PR [#4473](https://github.com/pipecat-ai/pipecat/pull/4473))
+
+### Changed
+
+- Updated the default `SonioxTTSService` model from `tts-rt-v1-preview` to the
+  generally available `tts-rt-v1`.
+  (PR [#4386](https://github.com/pipecat-ai/pipecat/pull/4386))
+
+- Default `cartesia_version` for `CartesiaTTSService` bumped from `2025-04-16`
+  to `2026-03-01`, matching `CartesiaHttpTTSService` and unlocking the
+  `use_normalized_timestamps` and `max_buffer_delay_ms` fields.
+  (PR [#4390](https://github.com/pipecat-ai/pipecat/pull/4390))
+
+- ⚠️ `CartesiaTTSService` now sends `use_normalized_timestamps: true` instead
+  of the deprecated `use_original_timestamps` field. Word timestamps now
+  reflect what was actually spoken (post text-normalization and
+  pronunciation-dictionary substitution), matching the convention Pipecat uses
+  for ElevenLabs. This is a behavior change for `sonic-3` users, who were
+  previously receiving timestamps tied to the input transcript.
+  (PR [#4390](https://github.com/pipecat-ai/pipecat/pull/4390))
+
+- Broadened `tool_resources` to `app_resources` for easy access not just in
+  tool handlers but in other places like custom `FrameProcessor`s. Three
+  changes: a rename (`tool_resources` → `app_resources`), a new `app_resources`
+  property on `PipelineTask`, and a new `pipeline_task` property on
+  `FrameProcessor`. Tool handlers now read `params.app_resources`; custom
+  processors read `self.pipeline_task.app_resources`. The previous
+  `tool_resources` aliases (on `PipelineTask`, `FunctionCallParams`, and
+  `FrameProcessorSetup`) keep working but are deprecated as of 1.2.0 and emit
+  `DeprecationWarning`s.
+  (PR [#4395](https://github.com/pipecat-ai/pipecat/pull/4395))
+
+- Lowered the per-message log in
+  `SmallWebRTCInputTransport._handle_app_message` from `debug` to `trace`. App
+  messages can be high-frequency and were noisy at debug level; set the loguru
+  level to `TRACE` to see them again.
+  (PR [#4397](https://github.com/pipecat-ai/pipecat/pull/4397))
+
+- Changed the default model for `GrokRealtimeLLMService` to
+  `grok-voice-think-fast-1.0`, xAI's recommended Voice Agent model. The
+  previous default of `grok-voice-fast-1.0` has been deprecated by xAI and is
+  being removed.
+  (PR [#4401](https://github.com/pipecat-ai/pipecat/pull/4401))
+
+- Changed the default Inworld TTS model from `inworld-tts-1.5-max` to
+  `inworld-tts-2` (Realtime TTS-2) across `InworldHttpTTSService`,
+  `InworldTTSService`, and the `InworldRealtimeLLMService` cascade. Existing
+  users can pin the prior model explicitly via the `model`/`tts_model`
+  argument; both `inworld-tts-1.5-max` and `inworld-tts-1.5-mini` remain valid
+  model IDs.
+  (PR [#4422](https://github.com/pipecat-ai/pipecat/pull/4422))
+
+- Changed the default model for `GrokLLMService` from `grok-3` to
+  `grok-4.20-non-reasoning`. xAI is retiring `grok-3` on May 15, 2026.
+  (PR [#4429](https://github.com/pipecat-ai/pipecat/pull/4429))
+
+- `DeepgramFluxSTT` watchdog silence threshold is now dynamic:
+  `max(chunk_duration * 2, watchdog_min_timeout)` instead of a fixed 500 ms.
+  This prevents false silence injections when large audio chunks are sent at
+  lower frequency.
+  (PR [#4430](https://github.com/pipecat-ai/pipecat/pull/4430))
+
+- `ElevenLabsTTSService` now sends `close_context` to the server as soon as the
+  turn is complete (on `on_turn_context_completed`) rather than waiting until
+  all audio has finished playing back. The `isFinal` message from ElevenLabs is
+  now used to signal `TTSStoppedFrame` and clean up the audio context,
+  improving turn transition timing.
+  (PR [#4433](https://github.com/pipecat-ai/pipecat/pull/4433))
+
+- Updated `InworldHttpTTSService` and `InworldTTSService` to use PCM audio
+  encoding by default, which returns audio bytes without headers.
+  (PR [#4446](https://github.com/pipecat-ai/pipecat/pull/4446))
+
+- Moved `create_task`, `cancel_task`, the `task_manager` property, and
+  `setup(task_manager)` up from `FrameProcessor` to `BaseObject`. Custom
+  `BaseObject` subclasses (turn strategies, controllers, etc.) now inherit
+  these methods directly instead of reimplementing the task manager wiring.
+  Owners propagate the task manager to their child `BaseObject`s via `await
+  child.setup(task_manager)`.
+  (PR [#4449](https://github.com/pipecat-ai/pipecat/pull/4449))
+
+- Changed the default OpenAI Realtime input audio transcription model from
+  `gpt-4o-transcribe` to `gpt-realtime-whisper` for both
+  `OpenAIRealtimeSTTService` and `OpenAIRealtimeLLMService`. The new model does
+  not accept the `prompt` parameter; if a prompt is supplied alongside
+  `gpt-realtime-whisper`, it is dropped automatically and a warning is logged.
+  To keep using prompt hints, explicitly pin `model="gpt-4o-transcribe"` (or
+  `"gpt-4o-mini-transcribe"`).
+  (PR [#4450](https://github.com/pipecat-ai/pipecat/pull/4450))
+
+- Updated the default model for `CartesiaTTSService` and
+  `CartesiaHttpTTSService` from `sonic-3` to `sonic-3.5`.
+  (PR [#4462](https://github.com/pipecat-ai/pipecat/pull/4462))
+
+- Changed the default model for `OpenAIRealtimeLLMService` from
+  `gpt-realtime-1.5` to `gpt-realtime-2`.
+  (PR [#4472](https://github.com/pipecat-ai/pipecat/pull/4472))
+
+### Deprecated
+
+- Deprecated `LLMUserAggregatorParams.filter_incomplete_user_turns`. Use
+  `user_turn_strategies=FilterIncompleteUserTurnStrategies()` (or add
+  `LLMTurnCompletionUserTurnStopStrategy` to a custom
+  `user_turn_strategies.stop`) instead. Setting the legacy flag still works for
+  one release: the aggregator emits a `DeprecationWarning` and rewires the
+  strategies as if you had passed `FilterIncompleteUserTurnStrategies`
+  directly.
+  (PR [#4405](https://github.com/pipecat-ai/pipecat/pull/4405))
+
+- Deprecated `ResampyResampler` in favor of `SOXRAudioResampler` (or the
+  `create_file_resampler()` / `create_stream_resampler()` factories).
+  Instantiating `ResampyResampler` now emits a `DeprecationWarning`. The class
+  will be removed in Pipecat 2.0 along with the default `resampy` and `numba`
+  dependencies.
+  (PR [#4428](https://github.com/pipecat-ai/pipecat/pull/4428))
+
+### Fixed
+
+- Fixed `CartesiaTTSService` surfacing `flush_done` messages from Cartesia as
+  `ErrorFrame`s. The latest API emits a `flush_done` per transcript when
+  server-side buffering is disabled; Pipecat now consumes them silently since
+  each turn already has its own `context_id`.
+  (PR [#4390](https://github.com/pipecat-ai/pipecat/pull/4390))
+
+- Fixed Cartesia tag helpers (`SPELL`, `EMOTION_TAG`, `PAUSE_TAG`,
+  `VOLUME_TAG`, `SPEED_TAG`) raising `TypeError` when called on an instance
+  (e.g. `tts.SPELL("hi")`). They're now `@staticmethod` and callable from both
+  the class and an instance.
+  (PR [#4390](https://github.com/pipecat-ai/pipecat/pull/4390))
+
+- Fixed `CartesiaHttpTTSService` pushing two `ErrorFrame`s on a non-200
+  response — one with the API's error text and a second, less informative
+  "Unknown error" frame from the outer exception handler. It now pushes a
+  single frame that includes the HTTP status code and returns cleanly.
+  (PR [#4390](https://github.com/pipecat-ai/pipecat/pull/4390))
+
+- Fixed an issue where `LocalSmartTurnAnalyzerV3` was imported unconditionally
+  for user turn stop strategies. It is now only imported when
+  `default_user_turn_stop_strategies()` is called. This improves startup time
+  and removes the `transformers` "PyTorch/TensorFlow/Flax not found" warning
+  when the default stop strategies are not used.
+  (PR [#4393](https://github.com/pipecat-ai/pipecat/pull/4393))
+
+- Fixed `GrokRealtimeLLMService` ignoring the configured model. The model was
+  stored in `Settings` but never sent to xAI, so every session silently fell
+  back to xAI's server-side default. The model is now passed via the `?model=`
+  query parameter on the WebSocket URL as xAI's Voice Agent API requires.
+  (PR [#4401](https://github.com/pipecat-ai/pipecat/pull/4401))
+
+- Fixed `on_user_turn_stopped` firing prematurely when
+  `filter_incomplete_user_turns` was enabled. The event now fires only after
+  the LLM confirms the user turn is complete (`✓`); previously the smart-turn
+  detector's tentative stop was bubbling up before the LLM had a chance to veto
+  it, causing observers, transcript appenders and UI indicators to receive an
+  early — and sometimes duplicated — signal.
+  (PR [#4405](https://github.com/pipecat-ai/pipecat/pull/4405))
+
+- Fixed `TTSSpeakFrame(append_to_context=True)` greetings sometimes splitting
+  across two assistant messages in the LLM context and not surfacing in
+  `on_assistant_turn_stopped`. The `LLMAssistantPushAggregationFrame` emitted
+  at the end of a TTS context now carries a PTS just past the last word so it
+  can't overtake clock-queued `TTSTextFrame`s in the transport's output, and
+  `LLMAssistantAggregator` now triggers
+  `on_assistant_turn_started`/`on_assistant_turn_stopped` when it receives the
+  frame outside an LLM response cycle (restoring v0.0.104 behavior for greeting
+  transcripts).
+  (PR [#4414](https://github.com/pipecat-ai/pipecat/pull/4414))
+
+- Fixed `ElevenLabsTTSService` and `ElevenLabsHttpTTSService` producing merged
+  words (e.g. `bookLook`) when using Flash models. Flash often splits sentences
+  mid-stream into alignment chunks that begin with a real inter-word space, but
+  the previous fix unconditionally stripped that space from every chunk.
+  Leading spaces are now stripped only on the first alignment chunk of an
+  utterance, so subsequent chunks correctly flush partial words across
+  boundaries.
+  (PR [#4415](https://github.com/pipecat-ai/pipecat/pull/4415))
+
+- Fixed AWS Polly TTS, Bedrock LLM, and the Bedrock AgentCore processor
+  erroring out when only one of `AWS_ACCESS_KEY_ID` / `AWS_SECRET_ACCESS_KEY`
+  was set in the environment. The half-populated kwargs are no longer forwarded
+  to aioboto3; partial env-var configurations now fall through to the boto3
+  credential chain like fully-unset configurations do.
+  (PR [#4416](https://github.com/pipecat-ai/pipecat/pull/4416))
+
+- Fixed `ElevenLabsTTSService` and `ElevenLabsHttpTTSService` writing
+  romanized/normalized text to the LLM context. With non-Latin input (e.g.,
+  Chinese), the assistant transcript was getting populated with pinyin (`Ni Hao
+  !` instead of `你好！`), which then degraded subsequent LLM turns. The services
+  now consume `alignment` by default and only switch to `normalizedAlignment` /
+  `normalized_alignment` when `pronunciation_dictionary_locators` is configured
+  (where `alignment` has overlapping restarts that produce duplicated/garbled
+  words, per #4316). Both fields are read with preferred-with-fallback
+  semantics since each is nullable per the API schema.
+  (PR [#4424](https://github.com/pipecat-ai/pipecat/pull/4424))
+
+- Fixed a deadlock in `TTSService` that could permanently stall pipeline
+  processing when all three conditions occurred together:
+  `pause_frame_processing=True`, an interruption arrived before any TTS audio
+  was played, and an `UninterruptibleFrame` (e.g. `TTSUpdateSettingsFrame`,
+  `FunctionCallResultFrame`) was in the processing queue at that moment. The
+  process task would block on `__process_event.wait()` indefinitely because
+  `BotStoppedSpeakingFrame` never arrives (no audio was played) and the
+  interruption handler did not resume processing. Affects services using
+  `pause_frame_processing=True` such as ElevenLabs, Rime, AsyncAI, Gradium, and
+  ResembleAI.
+  (PR [#4431](https://github.com/pipecat-ai/pipecat/pull/4431))
+
+- Fixed interruptions being delayed when a slow non-uninterruptible frame was
+  processing and an uninterruptible frame was waiting in the queue. The bot
+  would stall until the slow frame finished instead of cancelling it
+  immediately on interruption.
+  (PR [#4434](https://github.com/pipecat-ai/pipecat/pull/4434))
+
+- Fixed `TTSService` dropping uninterruptible frames (e.g.
+  `FunctionCallResultFrame`) from its internal serialization queue when an
+  interruption occurs. Previously, the queue was recreated on every
+  interruption, silently discarding any queued frames. The queue is now reset
+  instead of recreated, preserving uninterruptible frames so they are always
+  delivered downstream.
+  (PR [#4435](https://github.com/pipecat-ai/pipecat/pull/4435))
+
+- Fixed a race condition in the Daily transport that caused `AttributeError:
+  'NoneType' object has no attribute 'send_app_message'` when tearing down a
+  pipeline. Both `DailyInputTransport` and `DailyOutputTransport` share the
+  same `DailyTransportClient` and both call `cleanup()`, which was releasing
+  the underlying `CallClient` on the first call — leaving the second caller
+  with a `None` client.
+  (PR [#4440](https://github.com/pipecat-ai/pipecat/pull/4440))
+
+- Restored `cancel_on_interruption=False` support for `AWSNovaSonicLLMService`
+  and `OpenAIRealtimeLLMService`. These services previously honored the flag by
+  simply not cancelling in-flight function calls on interruption; the
+  introduction of the new async-tool mechanism (which threads
+  started/intermediate/final messages through the LLM context) broke that path
+  because the realtime services didn't know how to interpret those messages.
+  Note that new-style streamed intermediate results
+  (`FunctionCallResultProperties(is_final=False)`) are not supported on these
+  realtime services. Similar fixes for other impacted realtime services are
+  forthcoming.
+  (PR [#4441](https://github.com/pipecat-ai/pipecat/pull/4441))
+
+- Fixed two misspelled Gemini TTS voice names in
+  `GeminiTTSService.AVAILABLE_VOICES`.
+  (PR [#4443](https://github.com/pipecat-ai/pipecat/pull/4443))
+
+- Extended the `cancel_on_interruption=False` regression fix to
+  `GrokRealtimeLLMService`, `AzureRealtimeLLMService`, and
+  `UltravoxRealtimeLLMService`. Grok and Azure use the same approach as in
+  #4441 (each service detects async-tool messages in the LLM context and routes
+  the final result to its formal tool-result channel; Azure inherits
+  transitively from `OpenAIRealtimeLLMService`). Ultravox needed a different
+  approach because its API freezes the conversation between
+  `client_tool_invocation` and the matching `client_tool_result` — for
+  async-registered functions it now ships a placeholder `client_tool_result`
+  immediately when the function is invoked (to unfreeze the conversation), then
+  injects the real result as user-side text once the tool finishes. Streamed
+  intermediate results (`FunctionCallResultProperties(is_final=False)`) are
+  still not supported on any of these realtime services. `GeminiLiveLLMService`
+  and `InworldRealtimeLLMService` are excluded for now: Gemini Live's
+  async-tool path needs deeper investigation, and Inworld tool calling needs to
+  be sorted out first.
+  (PR [#4447](https://github.com/pipecat-ai/pipecat/pull/4447))
+
+- Fixed `OpenAIRealtimeLLMService` handling of multi-output-item responses
+  (observed with `gpt-realtime-2`). A single response can now contain more than
+  one audio item, and the first item's `audio.done` may arrive after the second
+  item's deltas have started. Deltas still arrive strictly in playback order,
+  so we continue to forward them as received (matching OpenAI's reference
+  implementation). The fix removes spurious warnings, ensures truncation always
+  targets the latest audio item, and emits a single bracketing
+  `TTSStartedFrame`/`TTSStoppedFrame` pair per assistant turn (the Stopped is
+  now pushed on `response.done`).
+  (PR [#4465](https://github.com/pipecat-ai/pipecat/pull/4465))
+
+- Fixed missing `output` attribute on LLM OpenTelemetry spans when the LLM call
+  is interrupted mid-stream.
+  (PR [#4467](https://github.com/pipecat-ai/pipecat/pull/4467))
+
+- Fixed incorrect `metrics.ttfb` on STT OpenTelemetry spans, and parented them
+  to the current turn span.
+  (PR [#4467](https://github.com/pipecat-ai/pipecat/pull/4467))
+
+- Fixed incorrect `metrics.ttfb` on TTS OpenTelemetry spans for streaming
+  services.
+  (PR [#4467](https://github.com/pipecat-ai/pipecat/pull/4467))
+
+- Extended the `cancel_on_interruption=False` regression fix to
+  `InworldRealtimeLLMService`. Uses the same approach as in #4441 (the service
+  detects async-tool messages in the LLM context and routes the final result to
+  its formal tool-result channel). Note: as of this writing, Inworld Realtime
+  doesn't appear to handle the resulting delayed tool result reliably — the
+  routing is best-effort and the service surfaces a one-time warning when
+  async-tool messages are seen. Streamed intermediate results
+  (`FunctionCallResultProperties(is_final=False)`) are still not supported on
+  this realtime service. (Inworld was excluded from #4447 pending resolution of
+  an unrelated tool-calling issue, which turned out to be an account-level
+  matter.)
+  (PR [#4474](https://github.com/pipecat-ai/pipecat/pull/4474))
+
+- Fixed Cartesia TTS Korean word timestamps to use normal spacing rules,
+  preserving word boundaries and per-word timestamp alignment during downstream
+  aggregation.
+  (PR [#4475](https://github.com/pipecat-ai/pipecat/pull/4475))
+
+- Fixed Cartesia TTS Chinese and Japanese timestamp grouping to preserve
+  provider text spacing, avoiding artificial spaces when timestamp groups are
+  reassembled downstream.
+  (PR [#4475](https://github.com/pipecat-ai/pipecat/pull/4475))
+
+- Fixed `SonioxSTTService` final transcription frames missing detected language
+  metadata when Soniox returns token-level language annotations.
+  (PR [#4482](https://github.com/pipecat-ai/pipecat/pull/4482))
+
+- Fixed Soniox final transcription language detection to use the most common
+  recognized token language, avoiding mislabeling an utterance when the last
+  token is tagged with a different language.
+  (PR [#4495](https://github.com/pipecat-ai/pipecat/pull/4495))
+
+- Fixed dropped audio in streaming TTS services whose wire protocol doesn't
+  echo `context_id` back on incoming audio (Sarvam, Smallest, Soniox, Inworld,
+  and others). Previously, audio that arrived between contexts or at the very
+  start of a turn was tagged with `context_id=None` and silently dropped with
+  an "unable to append audio to context: no context ID provided" debug log.
+  `TTSService.get_active_audio_context_id()` now falls back to the
+  synthesis-side `_turn_context_id` when the playback cursor isn't set yet.
+  (PR [#4497](https://github.com/pipecat-ai/pipecat/pull/4497))
+
+### Security
+
+- Fixed a path traversal issue in the development runner's
+  `/files/{filename:path}` download endpoint. Previously, when the runner was
+  started with `--folder`, a request like `/files/..%2F..%2Fetc%2Fpasswd` could
+  escape the configured folder because `%2F`-encoded separators bypassed
+  Starlette's path normalisation. The endpoint now resolves the joined path and
+  rejects any filename that escapes the allowed base with a 403, and also
+  returns 404 (instead of an implicit `null` 200) when `--folder` is unset.
+  (PR [#4417](https://github.com/pipecat-ai/pipecat/pull/4417))
+
 ## [1.1.0] - 2026-04-27
 
 ### Added

CLAUDE.md

@@ -1,157 +1 @@
-# CLAUDE.md
-
-This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
-
-## Project Overview
-
-Pipecat is an open-source Python framework for building real-time voice and multimodal conversational AI agents. It orchestrates audio/video, AI services, transports, and conversation pipelines using a frame-based architecture.
-
-## Common Commands
-
-```bash
-# Setup development environment
-uv sync --group dev --all-extras --no-extra gstreamer
-
-# Install pre-commit hooks
-uv run pre-commit install
-
-# Run all tests
-uv run pytest
-
-# Run a single test file
-uv run pytest tests/test_name.py
-
-# Run a specific test
-uv run pytest tests/test_name.py::test_function_name
-
-# Preview changelog
-uv run towncrier build --draft --version Unreleased
-
-# Lint and format check
-uv run ruff check
-uv run ruff format --check
-
-# Update dependencies (after editing pyproject.toml)
-uv lock && uv sync
-```
-
-## Architecture
-
-### Frame-Based Pipeline Processing
-
-All data flows as **Frame** objects through a pipeline of **FrameProcessors**:
-
-```
-[Processor1] → [Processor2] → ... → [ProcessorN]
-```
-
-**Key components:**
-
-- **Frames** (`src/pipecat/frames/frames.py`): Data units (audio, text, video) and control signals. Flow DOWNSTREAM (input→output) or UPSTREAM (acknowledgments/errors).
-
-- **FrameProcessor** (`src/pipecat/processors/frame_processor.py`): Base processing unit. Each processor receives frames, processes them, and pushes results downstream.
-
-- **Pipeline** (`src/pipecat/pipeline/pipeline.py`): Chains processors together.
-
-- **ParallelPipeline** (`src/pipecat/pipeline/parallel_pipeline.py`): Runs multiple pipelines in parallel.
-
-- **Transports** (`src/pipecat/transports/`): Transports are frame processors used for external I/O layer (Daily WebRTC, LiveKit WebRTC, WebSocket, Local). Abstract interface via `BaseTransport`, `BaseInputTransport` and `BaseOutputTransport`.
-
-- **Pipeline Task (`src/pipecat/pipeline/task.py`)**: Runs and manages a pipeline. Pipeline tasks send the first frame, `StartFrame`, to the pipeline in order for processors to know they can start processing and pushing frames. Pipeline tasks internally create a pipeline with two additional processors, a source processor before the user-defined pipeline and a sink processor at the end. Those are used for multiple things: error handling, pipeline task level events, heartbeat monitoring, etc.
-
-- **Pipeline Runner (`src/pipecat/pipeline/runner.py`)**: High-level entry point for executing pipeline tasks. Handles signal management (SIGINT/SIGTERM) for graceful shutdown and optional garbage collection. Run a single pipeline task with `await runner.run(task)` or multiple concurrently with `await asyncio.gather(runner.run(task1), runner.run(task2))`.
-
-- **Services** (`src/pipecat/services/`): 60+ AI provider integrations (STT, TTS, LLM, etc.). Extend base classes: `AIService`, `LLMService`, `STTService`, `TTSService`, `VisionService`.
-
-- **Serializers** (`src/pipecat/serializers/`): Convert frames to/from wire formats for WebSocket transports. `FrameSerializer` base class defines `serialize()` and `deserialize()`. Telephony serializers (Twilio, Plivo, Vonage, Telnyx, Exotel, Genesys) handle provider-specific protocols and audio encoding (e.g., μ-law).
-
-- **RTVI** (`src/pipecat/processors/frameworks/rtvi.py`): Real-Time Voice Interface protocol bridging clients and the pipeline. `RTVIProcessor` handles incoming client messages (text input, audio, function call results). `RTVIObserver` converts pipeline frames to outgoing messages: user/bot speaking events, transcriptions, LLM/TTS lifecycle, function calls, metrics, and audio levels.
-
-- **Observers** (`src/pipecat/observers/`): Monitor frame flow without modifying the pipeline. Passed to `PipelineTask` via the `observers` parameter. Implement `on_process_frame()` and `on_push_frame()` callbacks.
-
-### Important Patterns
-
-- **Context Aggregation**: `LLMContext` accumulates messages for LLM calls; `UserResponse` aggregates user input
-
-- **Turn Management**: Turn management is done through `LLMUserAggregator` and
-  `LLMAssistantAggregator`, created with `LLMContextAggregatorPair`
-
-- **User turn strategies**: Detection of when the user starts and stops speaking is done via user turn start/stop strategies. They push `UserStartedSpeakingFrame` and `UserStoppedSpeakingFrame` respectively.
-
-- **Interruptions**: Interruptions are usually triggered by a user turn start strategy (e.g. `VADUserTurnStartStrategy`) but they can be triggered by other processors as well, in which case the user turn start strategies don't need to. An `InterruptionFrame` carries an optional `asyncio.Event` that is set when the frame reaches the pipeline sink. If a processor stops an `InterruptionFrame` from propagating downstream (i.e., doesn't push it), it **must** call `frame.complete()` to avoid stalling `push_interruption_task_frame_and_wait()` callers.
-
-- **Uninterruptible Frames**: These are frames that will not be removed from internal queues even if there's an interruption. For example, `EndFrame` and `StopFrame`.
-
-- **Events**: Most classes in Pipecat have `BaseObject` as the very base class. `BaseObject` has support for events. Events can run in the background in an async task (default) or synchronously (`sync=True`) if we want immediate action. Synchronous event handlers need to execute fast.
-
-- **Async Task Management**: Always use `self.create_task(coroutine, name)` instead of raw `asyncio.create_task()`. The `TaskManager` automatically tracks tasks and cleans them up on processor shutdown. Use `await self.cancel_task(task, timeout)` for cancellation.
-
-- **Error Handling**: Use `await self.push_error(msg, exception, fatal)` to push errors upstream. Services should use `fatal=False` (the default) so application code can handle errors and take action (e.g. switch to another service).
-
-### Key Directories
-
-| Directory                  | Purpose                                            |
-| -------------------------- | -------------------------------------------------- |
-| `src/pipecat/frames/`      | Frame definitions (100+ types)                     |
-| `src/pipecat/processors/`  | FrameProcessor base + aggregators, filters, audio  |
-| `src/pipecat/pipeline/`    | Pipeline orchestration                             |
-| `src/pipecat/services/`    | AI service integrations (60+ providers)            |
-| `src/pipecat/transports/`  | Transport layer (Daily, LiveKit, WebSocket, Local) |
-| `src/pipecat/serializers/` | Frame serialization for WebSocket protocols        |
-| `src/pipecat/observers/`   | Pipeline observers for monitoring frame flow       |
-| `src/pipecat/audio/`       | VAD, filters, mixers, turn detection, DTMF         |
-| `src/pipecat/turns/`       | User turn management                               |
-
-## Code Style
-
-- **Docstrings**: Google-style. Classes describe purpose; `__init__` has `Args:` section; dataclasses use `Parameters:` section.
-- **Linting**: Ruff (line length 100). Pre-commit hooks enforce formatting.
-- **Type hints**: Required for complex async code.
-- **Dataclass vs Pydantic**: Use `@dataclass` for frames and internal pipeline data (high-frequency, no validation needed). Use Pydantic `BaseModel` for configuration, parameters, metrics, and external API data (benefits from validation and serialization). Specifically:
-  - `@dataclass`: Frame types, context aggregator pairs, internal data containers
-  - `BaseModel`: Service `InputParams`, transport/VAD/turn params, metrics data, API request/response models, serializer params
-
-### Docstring Example
-
-```python
-class MyService(LLMService):
-    """Description of what the service does.
-
-    More detailed description.
-
-    Event handlers available:
-
-    - on_connected: Called when we are connected
-
-    Example::
-
-        @service.event_handler("on_connected")
-        async def on_connected(service, frame):
-            ...
-    """
-
-    def __init__(self, param1: str, **kwargs):
-        """Initialize the service.
-
-        Args:
-            param1: Description of param1.
-            **kwargs: Additional arguments passed to parent.
-        """
-        super().__init__(**kwargs)
-```
-
-## Service Implementation
-
-When adding a new service:
-
-1. Extend the appropriate base class (`STTService`, `TTSService`, `LLMService`, etc.)
-2. Implement required abstract methods
-3. Handle necessary frames
-4. By default, all frames should be pushed in the direction they came
-5. Push `ErrorFrame` on failures
-6. Add metrics tracking via `MetricsData` if relevant
-7. Follow the pattern of existing services in `src/pipecat/services/`
-
-## Testing
-
-Test utilities live in `src/pipecat/tests/utils.py`. Use `run_test()` to send frames through a pipeline and assert expected output frames in each direction. Use `SleepFrame(sleep=N)` to add delays between frames.
+@AGENTS.md

README.md

@@ -92,10 +92,10 @@ Catch new features, interviews, and how-tos on our [Pipecat TV](https://www.yout
 | Category            | Services                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                            |
 | ------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | Speech-to-Text      | [AssemblyAI](https://docs.pipecat.ai/api-reference/server/services/stt/assemblyai), [AWS](https://docs.pipecat.ai/api-reference/server/services/stt/aws), [Azure](https://docs.pipecat.ai/api-reference/server/services/stt/azure), [Cartesia](https://docs.pipecat.ai/api-reference/server/services/stt/cartesia), [Deepgram](https://docs.pipecat.ai/api-reference/server/services/stt/deepgram), [ElevenLabs](https://docs.pipecat.ai/api-reference/server/services/stt/elevenlabs), [Fal Wizper](https://docs.pipecat.ai/api-reference/server/services/stt/fal), [Gladia](https://docs.pipecat.ai/api-reference/server/services/stt/gladia), [Google](https://docs.pipecat.ai/api-reference/server/services/stt/google), [Gradium](https://docs.pipecat.ai/api-reference/server/services/stt/gradium), [Groq (Whisper)](https://docs.pipecat.ai/api-reference/server/services/stt/groq), [Mistral](https://docs.pipecat.ai/api-reference/server/services/stt/mistral), [NVIDIA](https://docs.pipecat.ai/api-reference/server/services/stt/nvidia), [OpenAI (Whisper)](https://docs.pipecat.ai/api-reference/server/services/stt/openai), [Sarvam](https://docs.pipecat.ai/api-reference/server/services/stt/sarvam), [Soniox](https://docs.pipecat.ai/api-reference/server/services/stt/soniox), [Speechmatics](https://docs.pipecat.ai/api-reference/server/services/stt/speechmatics), [Whisper](https://docs.pipecat.ai/api-reference/server/services/stt/whisper), [xAI](https://docs.pipecat.ai/api-reference/server/services/stt/xai)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     |
-| LLMs                | [Anthropic](https://docs.pipecat.ai/api-reference/server/services/llm/anthropic), [AWS](https://docs.pipecat.ai/api-reference/server/services/llm/aws), [Azure](https://docs.pipecat.ai/api-reference/server/services/llm/azure), [Cerebras](https://docs.pipecat.ai/api-reference/server/services/llm/cerebras), [DeepSeek](https://docs.pipecat.ai/api-reference/server/services/llm/deepseek), [Fireworks AI](https://docs.pipecat.ai/api-reference/server/services/llm/fireworks), [Gemini](https://docs.pipecat.ai/api-reference/server/services/llm/gemini), [Grok](https://docs.pipecat.ai/api-reference/server/services/llm/grok), [Groq](https://docs.pipecat.ai/api-reference/server/services/llm/groq), [Mistral](https://docs.pipecat.ai/api-reference/server/services/llm/mistral), [Nebius](https://docs.pipecat.ai/api-reference/server/services/llm/nebius), [Novita](https://docs.pipecat.ai/api-reference/server/services/llm/novita), [NVIDIA NIM](https://docs.pipecat.ai/api-reference/server/services/llm/nvidia), [Ollama](https://docs.pipecat.ai/api-reference/server/services/llm/ollama), [OpenAI](https://docs.pipecat.ai/api-reference/server/services/llm/openai), [OpenAI Responses](https://docs.pipecat.ai/api-reference/server/services/llm/openai-responses), [OpenRouter](https://docs.pipecat.ai/api-reference/server/services/llm/openrouter), [Perplexity](https://docs.pipecat.ai/api-reference/server/services/llm/perplexity), [Qwen](https://docs.pipecat.ai/api-reference/server/services/llm/qwen), [SambaNova](https://docs.pipecat.ai/api-reference/server/services/llm/sambanova), [Sarvam](https://docs.pipecat.ai/api-reference/server/services/llm/sarvam), [Together AI](https://docs.pipecat.ai/api-reference/server/services/llm/together)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    |
+| LLMs                | [Anthropic](https://docs.pipecat.ai/api-reference/server/services/llm/anthropic), [AWS](https://docs.pipecat.ai/api-reference/server/services/llm/aws), [Azure](https://docs.pipecat.ai/api-reference/server/services/llm/azure), [Cerebras](https://docs.pipecat.ai/api-reference/server/services/llm/cerebras), [DeepSeek](https://docs.pipecat.ai/api-reference/server/services/llm/deepseek), [Fireworks AI](https://docs.pipecat.ai/api-reference/server/services/llm/fireworks), [Gemini](https://docs.pipecat.ai/api-reference/server/services/llm/gemini), [Grok](https://docs.pipecat.ai/api-reference/server/services/llm/grok), [Groq](https://docs.pipecat.ai/api-reference/server/services/llm/groq), [Inception](https://docs.pipecat.ai/api-reference/server/services/llm/inception), [Mistral](https://docs.pipecat.ai/api-reference/server/services/llm/mistral), [Nebius](https://docs.pipecat.ai/api-reference/server/services/llm/nebius), [Novita](https://docs.pipecat.ai/api-reference/server/services/llm/novita), [NVIDIA NIM](https://docs.pipecat.ai/api-reference/server/services/llm/nvidia), [Ollama](https://docs.pipecat.ai/api-reference/server/services/llm/ollama), [OpenAI](https://docs.pipecat.ai/api-reference/server/services/llm/openai), [OpenAI Responses](https://docs.pipecat.ai/api-reference/server/services/llm/openai-responses), [OpenRouter](https://docs.pipecat.ai/api-reference/server/services/llm/openrouter), [Perplexity](https://docs.pipecat.ai/api-reference/server/services/llm/perplexity), [Qwen](https://docs.pipecat.ai/api-reference/server/services/llm/qwen), [SambaNova](https://docs.pipecat.ai/api-reference/server/services/llm/sambanova), [Sarvam](https://docs.pipecat.ai/api-reference/server/services/llm/sarvam), [Together AI](https://docs.pipecat.ai/api-reference/server/services/llm/together)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    |
 | Text-to-Speech      | [Async](https://docs.pipecat.ai/api-reference/server/services/tts/asyncai), [AWS](https://docs.pipecat.ai/api-reference/server/services/tts/aws), [Azure](https://docs.pipecat.ai/api-reference/server/services/tts/azure), [Camb AI](https://docs.pipecat.ai/api-reference/server/services/tts/camb), [Cartesia](https://docs.pipecat.ai/api-reference/server/services/tts/cartesia), [Deepgram](https://docs.pipecat.ai/api-reference/server/services/tts/deepgram), [ElevenLabs](https://docs.pipecat.ai/api-reference/server/services/tts/elevenlabs), [Fish](https://docs.pipecat.ai/api-reference/server/services/tts/fish), [Google](https://docs.pipecat.ai/api-reference/server/services/tts/google), [Gradium](https://docs.pipecat.ai/api-reference/server/services/tts/gradium), [Groq](https://docs.pipecat.ai/api-reference/server/services/tts/groq), [Hume](https://docs.pipecat.ai/api-reference/server/services/tts/hume), [Inworld](https://docs.pipecat.ai/api-reference/server/services/tts/inworld), [Kokoro](https://docs.pipecat.ai/api-reference/server/services/tts/kokoro), [LMNT](https://docs.pipecat.ai/api-reference/server/services/tts/lmnt), [MiniMax](https://docs.pipecat.ai/api-reference/server/services/tts/minimax), [Mistral](https://docs.pipecat.ai/api-reference/server/services/tts/mistral), [Neuphonic](https://docs.pipecat.ai/api-reference/server/services/tts/neuphonic), [NVIDIA](https://docs.pipecat.ai/api-reference/server/services/tts/nvidia), [OpenAI](https://docs.pipecat.ai/api-reference/server/services/tts/openai), [Piper](https://docs.pipecat.ai/api-reference/server/services/tts/piper), [Resemble](https://docs.pipecat.ai/api-reference/server/services/tts/resemble), [Rime](https://docs.pipecat.ai/api-reference/server/services/tts/rime), [Sarvam](https://docs.pipecat.ai/api-reference/server/services/tts/sarvam), [Smallest](https://docs.pipecat.ai/api-reference/server/services/tts/smallest), [Soniox](https://docs.pipecat.ai/api-reference/server/services/tts/soniox), [Speechmatics](https://docs.pipecat.ai/api-reference/server/services/tts/speechmatics), [xAI](https://docs.pipecat.ai/api-reference/server/services/tts/xai), [XTTS](https://docs.pipecat.ai/api-reference/server/services/tts/xtts) |
 | Speech-to-Speech    | [AWS Nova Sonic](https://docs.pipecat.ai/api-reference/server/services/s2s/aws), [Gemini Multimodal Live](https://docs.pipecat.ai/api-reference/server/services/s2s/gemini), [Grok Voice Agent](https://docs.pipecat.ai/api-reference/server/services/s2s/grok), [OpenAI Realtime](https://docs.pipecat.ai/api-reference/server/services/s2s/openai), [Ultravox](https://docs.pipecat.ai/api-reference/server/services/s2s/ultravox),                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |
-| Transport           | [Daily (WebRTC)](https://docs.pipecat.ai/api-reference/server/services/transport/daily), [FastAPI Websocket](https://docs.pipecat.ai/api-reference/server/services/transport/fastapi-websocket), [LiveKit (WebRTC)](https://docs.pipecat.ai/api-reference/server/services/transport/livekit), [SmallWebRTCTransport](https://docs.pipecat.ai/api-reference/server/services/transport/small-webrtc), [WebSocket Server](https://docs.pipecat.ai/api-reference/server/services/transport/websocket-server), [WhatsApp](https://docs.pipecat.ai/api-reference/server/services/transport/whatsapp), Local                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |
+| Transport           | [Daily (WebRTC)](https://docs.pipecat.ai/api-reference/server/services/transport/daily), [FastAPI Websocket](https://docs.pipecat.ai/api-reference/server/services/transport/fastapi-websocket), [LiveKit (WebRTC)](https://docs.pipecat.ai/api-reference/server/services/transport/livekit), [SmallWebRTCTransport](https://docs.pipecat.ai/api-reference/server/services/transport/small-webrtc), [Vonage (WebRTC)](https://docs.pipecat.ai/api-reference/server/services/transport/vonage), [WebSocket Server](https://docs.pipecat.ai/api-reference/server/services/transport/websocket-server), [WhatsApp](https://docs.pipecat.ai/api-reference/server/services/transport/whatsapp), Local                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |
 | Serializers         | [Exotel](https://docs.pipecat.ai/api-reference/server/services/serializers/exotel), [Genesys](https://docs.pipecat.ai/api-reference/server/services/serializers/genesys), [Plivo](https://docs.pipecat.ai/api-reference/server/services/serializers/plivo), [Twilio](https://docs.pipecat.ai/api-reference/server/services/serializers/twilio), [Telnyx](https://docs.pipecat.ai/api-reference/server/services/serializers/telnyx), [Vonage](https://docs.pipecat.ai/api-reference/server/services/serializers/vonage)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                              |
 | Video               | [HeyGen](https://docs.pipecat.ai/api-reference/server/services/video/heygen), [LemonSlice](https://docs.pipecat.ai/api-reference/server/services/transport/lemonslice), [Tavus](https://docs.pipecat.ai/api-reference/server/services/video/tavus), [Simli](https://docs.pipecat.ai/api-reference/server/services/video/simli)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      |
 | Memory              | [mem0](https://docs.pipecat.ai/api-reference/server/services/memory/mem0)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           |

changelog/4052.added.md

@@ -0,0 +1 @@
+- Added `VonageVideoConnectorTransport`, a new transport integration for real-time Vonage WebRTC sessions using the Vonage Video Connector library.

changelog/4306.fixed.md

@@ -0,0 +1 @@
+- Fixed Azure TTS last word being missed by observers and RTVI UI. The completion signal was racing with word timestamp processing, causing the final word's `TTSTextFrame` to arrive after `TTSStoppedFrame`. Completion is now routed through the word boundary queue to ensure all words are processed before signaling stream end.

changelog/4380.fixed.2.md

@@ -0,0 +1 @@
+- Fixed `BaseOutputTransport` reordering frames that share the same presentation timestamp. Frames with equal PTS values are now emitted in insertion order, preventing subtle audio/text sequencing bugs when multiple frames arrive at the same time.

changelog/4380.fixed.3.md

@@ -0,0 +1 @@
+- Fixed Cartesia word timestamps leaking SSML tag text (e.g. `<spell>`, `<emotion>`, `<break>`) into word entries. Tags are now stripped before processing, so word-to-text attribution remains accurate when SSML markup is present in the TTS input.

changelog/4380.fixed.4.md

@@ -0,0 +1 @@
+- Fixed `TTSTextFrame` entries losing their original text structure when word timestamps are enabled. Each `TTSTextFrame` now carries a `raw_text` field containing the corresponding span of the original LLM-produced text (including pattern delimiters such as `<card>4111 1111 1111 1111</card>`), so the assistant context receives properly-tagged content rather than the cleaned words returned by the TTS provider. Also handles words that straddle two sentence boundaries by splitting them and attributing each part to its correct source frame.

changelog/4380.fixed.md

@@ -0,0 +1 @@
+- Fixed skipped TTS frames (e.g. code blocks filtered via `skip_aggregator_types`) being emitted to the assistant context immediately instead of waiting for preceding spoken frames to finish. They now hold their position in the frame sequence and are flushed only after all earlier spoken sentences are complete, keeping context ordering correct.

changelog/4385.added.md

@@ -1 +0,0 @@
-- Added a `session_id` field to `RunnerArguments` so bots can log or trace a per-session identifier in local development the same way they can in Pipecat Cloud. The development runner now mints a UUID at every construction site, and paths that already returned a `sessionId` to the caller (Daily `/start`, dial-in webhook) share that same UUID with the runner args instead of generating two. The SmallWebRTC `/api/offer` endpoint also accepts an optional `session_id` query parameter so the `/sessions/{session_id}/...` proxy can thread it through.

changelog/4386.changed.md

@@ -1 +0,0 @@
-- Updated the default `SonioxTTSService` model from `tts-rt-v1-preview` to the generally available `tts-rt-v1`.

changelog/4390.added.md

@@ -1 +0,0 @@
-- Added a `max_buffer_delay_ms` constructor argument to `CartesiaTTSService` for controlling Cartesia's server-side text buffering. When unset, Pipecat picks a sensible default based on `text_aggregation_mode`: `0` in `SENTENCE` mode (custom buffering — avoids stacking client-side aggregation on top of Cartesia's default 3000ms server buffer) and unset in `TOKEN` mode (Cartesia's managed buffering applies). Pass an explicit value (0–5000ms) to override.

changelog/4390.changed.2.md

@@ -1 +0,0 @@
-- Default `cartesia_version` for `CartesiaTTSService` bumped from `2025-04-16` to `2026-03-01`, matching `CartesiaHttpTTSService` and unlocking the `use_normalized_timestamps` and `max_buffer_delay_ms` fields.

changelog/4390.changed.md

@@ -1 +0,0 @@
-- ⚠️ `CartesiaTTSService` now sends `use_normalized_timestamps: true` instead of the deprecated `use_original_timestamps` field. Word timestamps now reflect what was actually spoken (post text-normalization and pronunciation-dictionary substitution), matching the convention Pipecat uses for ElevenLabs. This is a behavior change for `sonic-3` users, who were previously receiving timestamps tied to the input transcript.

changelog/4390.fixed.2.md

@@ -1 +0,0 @@
-- Fixed `CartesiaHttpTTSService` pushing two `ErrorFrame`s on a non-200 response — one with the API's error text and a second, less informative "Unknown error" frame from the outer exception handler. It now pushes a single frame that includes the HTTP status code and returns cleanly.

changelog/4390.fixed.3.md

@@ -1 +0,0 @@
-- Fixed Cartesia tag helpers (`SPELL`, `EMOTION_TAG`, `PAUSE_TAG`, `VOLUME_TAG`, `SPEED_TAG`) raising `TypeError` when called on an instance (e.g. `tts.SPELL("hi")`). They're now `@staticmethod` and callable from both the class and an instance.

changelog/4390.fixed.md

@@ -1 +0,0 @@
-- Fixed `CartesiaTTSService` surfacing `flush_done` messages from Cartesia as `ErrorFrame`s. The latest API emits a `flush_done` per transcript when server-side buffering is disabled; Pipecat now consumes them silently since each turn already has its own `context_id`.

changelog/4393.fixed.md

@@ -1 +0,0 @@
-- Fixed an issue where `LocalSmartTurnAnalyzerV3` was imported unconditionally for user turn stop strategies. It is now only imported when `default_user_turn_stop_strategies()` is called. This improves startup time and removes the `transformers` "PyTorch/TensorFlow/Flax not found" warning when the default stop strategies are not used.

changelog/4395.changed.md

@@ -1 +0,0 @@
-- Broadened `tool_resources` to `app_resources` for easy access not just in tool handlers but in other places like custom `FrameProcessor`s. Three changes: a rename (`tool_resources` → `app_resources`), a new `app_resources` property on `PipelineTask`, and a new `pipeline_task` property on `FrameProcessor`. Tool handlers now read `params.app_resources`; custom processors read `self.pipeline_task.app_resources`. The previous `tool_resources` aliases (on `PipelineTask`, `FunctionCallParams`, and `FrameProcessorSetup`) keep working but are deprecated as of 1.2.0 and emit `DeprecationWarning`s.

changelog/4423.added.md

@@ -0,0 +1 @@
+- Added `InceptionLLMService` for Inception's Mercury 2 diffusion reasoning model, with support for `reasoning_effort` and `realtime` settings.

changelog/4442.added.2.md

@@ -0,0 +1 @@
+- Added `GET /status` endpoint to the development runner that reports which transports the running instance accepts (all by default, or the single transport passed via `-t`).

changelog/4442.added.md

@@ -0,0 +1 @@
+- Added plain WebSocket transport support to the development runner. Bots can now accept connections from non-telephony WebSocket clients (e.g., browser apps using protobuf framing) via the `/ws-client` endpoint alongside other transports.

changelog/4442.changed.md

@@ -0,0 +1 @@
+- ⚠️ The development runner now supports all transports (WebRTC, Daily, telephony, plain WebSocket) simultaneously from a single server. The `/start` endpoint accepts a `"transport"` field to select the transport per-request; omitting `-t` at startup enables all transports instead of defaulting to WebRTC. The Daily browser-redirect route moved from `GET /` to `GET /daily`.

changelog/4507.fixed.md

@@ -0,0 +1 @@
+- Fixed `ElevenLabsSTTService` crashing when `language` was passed as `None`. When `language` is not set, the service now lets ElevenLabs auto-detect the audio language.

changelog/4514.fixed.md

@@ -0,0 +1 @@
+- Fixed websocket STT connection setup failures so services clear stale websocket state and emit non-fatal error frames, allowing `ServiceSwitcher` failover to keep agents running.

changelog/4521.added.md

@@ -0,0 +1 @@
+- Added `max_endpoint_delay_ms` to `SonioxSTTService.Settings`, controlling the maximum delay (500-3000 ms) before endpoint detection finalizes a turn.

changelog/4521.changed.md

@@ -0,0 +1 @@
+- `SonioxSTTService` now applies settings updates (e.g. via `STTUpdateSettingsFrame`) using a graceful reconnect instead of a hard disconnect/reconnect, preserving the service's reconnect retry behavior.

changelog/4521.removed.md

@@ -0,0 +1 @@
+- Removed the unsupported Georgian (`Language.KA`) language mapping from `SonioxSTTService`.

changelog/4522.changed.md

@@ -0,0 +1 @@
+- Updated the default p99 TTFS latency values for Smallest AI, Mistral, and XAI STT so turn stop timing uses measured values instead of the conservative fallback.

changelog/4524.changed.md

@@ -0,0 +1 @@
+- Updated the development runner startup banner to show the prebuilt client URL once and list enabled or disabled transports with install hints.

changelog/4524.fixed.md

@@ -0,0 +1 @@
+- Fixed the development runner so missing optional transport dependencies disable only their related routes instead of failing startup in all-transport mode.

changelog/4527.fixed.md

@@ -0,0 +1 @@
+- Fixed a race in `ElevenLabsTTSService` where the periodic keepalive could be sent for a new turn's context before that context's `voice_settings` initialization message, causing ElevenLabs to close the WebSocket with a 1008 policy violation (`voice_settings field must be provided in the first message ...`). The keepalive now only targets a context once its context-init has been sent.

changelog/4531.changed.md

@@ -0,0 +1 @@
+- Bumped `pipecat-ai-prebuilt` to 1.0.1 in the `runner` extra, updating the prebuilt client UI served by the development runner.

env.example

@@ -91,6 +91,9 @@ HEYGEN_LIVE_AVATAR_API_KEY=...
 HUME_API_KEY=...
 HUME_VOICE_ID=...
 
+# Inception
+INCEPTION_API_KEY=...
+
 # Inworld
 INWORLD_API_KEY=...
 
@@ -132,6 +135,10 @@ NOVITA_API_KEY=...
 
 # NVIDIA
 NVIDIA_API_KEY=...
+# For a full example of how to deploy to SageMaker, see:
+# https://github.com/pipecat-ai/pipecat-examples/tree/main/nvidia_sagemaker_example/deployment/aws-sagemaker-nvidia
+SAGEMAKER_ASR_ENDPOINT_NAME=...
+SAGEMAKER_MAGPIE_ENDPOINT_NAME=...
 
 # OpenAI
 OPENAI_API_KEY=...
@@ -207,6 +214,11 @@ TWILIO_AUTH_TOKEN=...
 # Ultravox Realtime
 ULTRAVOX_API_KEY=...
 
+# Vonage
+VONAGE_APPLICATION_ID=...
+VONAGE_SESSION_ID=...
+VONAGE_TOKEN=...
+
 # WhatsApp
 WHATSAPP_TOKEN=...
 WHATSAPP_WEBHOOK_VERIFICATION_TOKEN=...

examples/features/features-add-tool-change-messages.py

@@ -0,0 +1,232 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Manual validation harness for the ``add_tool_change_messages`` feature.
+
+When tools change mid-conversation, LLMs can produce a few different
+flavors of tool-call-related hallucination:
+
+- **Forward hallucination** — calling a tool that has been removed.
+- **Negative hallucination** — refusing to call a tool that has been
+  re-added (because recent context is full of "I can't" responses).
+- **Hallucinated output when tools are unavailable** — making up an
+  answer rather than declining gracefully, or producing JSON that
+  *looks* like a tool call but is actually just an assistant text
+  response.
+
+The ``add_tool_change_messages`` feature mitigates these by appending a
+developer-role message to the conversation whenever ``LLMSetToolsFrame``
+changes the set of advertised tools, so the LLM stays in sync with what's
+actually available.
+
+This harness exercises all of those flavors by flipping the advertised
+tool set on a turn counter:
+
+    Phase 0 (turns 1–4):   weather tool ACTIVE — confirm baseline.
+    Phase 1 (turns 5–8):   tool REMOVED — keep asking for weather.
+    Phase 2 (turn 9+):     tool RE-ADDED — does the LLM call it again?
+
+Set ``ADD_TOOL_CHANGE_MESSAGES=0`` to disable the mitigation and see the
+unmitigated behavior. The default is ON so a fresh run shows the feature
+working.
+
+Defaults to Llama 3.1 8B Instruct via a locally-running Ollama —
+anecdotally one of the more hallucination-prone of the easily accessible
+models. Pull the model once with ``ollama pull llama3.1:8b`` and make
+sure ``ollama serve`` is running. Swap the LLM service to validate other
+providers.
+
+Run with::
+
+    uv run examples/features/features-add-tool-change-messages.py
+    ADD_TOOL_CHANGE_MESSAGES=0 uv run examples/features/features-add-tool-change-messages.py
+"""
+
+import os
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.adapters.schemas.function_schema import FunctionSchema
+from pipecat.adapters.schemas.tools_schema import ToolsSchema
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.frames.frames import LLMRunFrame, LLMSetToolsFrame
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.llm_context import NOT_GIVEN, LLMContext
+from pipecat.processors.aggregators.llm_response_universal import (
+    LLMContextAggregatorPair,
+    LLMUserAggregatorParams,
+)
+from pipecat.runner.types import RunnerArguments
+from pipecat.runner.utils import create_transport
+from pipecat.services.cartesia.tts import CartesiaTTSService
+from pipecat.services.deepgram.stt import DeepgramSTTService
+from pipecat.services.llm_service import FunctionCallParams
+from pipecat.services.ollama.llm import OLLamaLLMService
+from pipecat.transports.base_transport import BaseTransport, TransportParams
+from pipecat.transports.daily.transport import DailyParams
+from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
+
+load_dotenv(override=True)
+
+# Default ON so a fresh run shows the feature working. Set to "0" to A/B
+# against the unmitigated behavior.
+ADD_TOOL_CHANGE_MESSAGES = os.environ.get("ADD_TOOL_CHANGE_MESSAGES", "1") == "1"
+
+
+async def fetch_weather_from_api(params: FunctionCallParams):
+    await params.result_callback({"conditions": "nice", "temperature": "75"})
+
+
+weather_function = FunctionSchema(
+    name="get_current_weather",
+    description="Get the current weather",
+    properties={
+        "location": {
+            "type": "string",
+            "description": "The city and state, e.g. San Francisco, CA",
+        },
+        "format": {
+            "type": "string",
+            "enum": ["celsius", "fahrenheit"],
+            "description": "The temperature unit to use. Infer this from the user's location.",
+        },
+    },
+    required=["location", "format"],
+)
+weather_tools = ToolsSchema(standard_tools=[weather_function])
+
+
+transport_params = {
+    "daily": lambda: DailyParams(audio_in_enabled=True, audio_out_enabled=True),
+    "twilio": lambda: FastAPIWebsocketParams(audio_in_enabled=True, audio_out_enabled=True),
+    "webrtc": lambda: TransportParams(audio_in_enabled=True, audio_out_enabled=True),
+}
+
+
+async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
+    logger.info(
+        f"Starting add_tool_change_messages demo bot "
+        f"(ADD_TOOL_CHANGE_MESSAGES={ADD_TOOL_CHANGE_MESSAGES})"
+    )
+
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
+
+    tts = CartesiaTTSService(
+        api_key=os.environ["CARTESIA_API_KEY"],
+        settings=CartesiaTTSService.Settings(
+            voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
+        ),
+    )
+
+    llm = OLLamaLLMService(
+        settings=OLLamaLLMService.Settings(
+            # Llama 3.1 8B Instruct is anecdotally one of the more
+            # hallucination-prone of the easily accessible models — exactly
+            # what we want for this validation harness. Pull it with
+            # ``ollama pull llama3.1:8b`` and make sure ``ollama serve``
+            # is running.
+            model="llama3.1:8b",
+            system_instruction=(
+                "You are a helpful assistant in a voice conversation. Your responses "
+                "will be spoken aloud, so avoid emojis, bullet points, or other "
+                "formatting that can't be spoken. Respond briefly and naturally. "
+                "If the user asks for the current weather, use the `get_current_weather` "
+                "function if it's available. IMPORTANT: if you do not have access to the function, "
+                "say something along the lines of 'Sorry, I can't check the weather right now.'."
+            ),
+        ),
+    )
+    llm.register_function("get_current_weather", fetch_weather_from_api)
+
+    context = LLMContext(tools=weather_tools)
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
+        context,
+        user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()),
+        add_tool_change_messages=ADD_TOOL_CHANGE_MESSAGES,
+    )
+
+    pipeline = Pipeline(
+        [
+            transport.input(),
+            stt,
+            user_aggregator,
+            llm,
+            tts,
+            transport.output(),
+            assistant_aggregator,
+        ]
+    )
+
+    task = PipelineTask(
+        pipeline,
+        params=PipelineParams(enable_metrics=True, enable_usage_metrics=True),
+        idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
+    )
+
+    # Phase controller: roughly 4 turns per phase.
+    user_turn_count = 0
+    REMOVE_AT_TURN = 5  # tool gone for turn N onward
+    READD_AT_TURN = 9  # tool back for turn N onward
+
+    @user_aggregator.event_handler("on_user_turn_stopped")
+    async def on_user_turn_stopped(aggregator, strategy, message):
+        nonlocal user_turn_count
+        user_turn_count += 1
+        logger.info(f"=== User turn {user_turn_count} complete ===")
+
+        if user_turn_count == REMOVE_AT_TURN - 1:
+            logger.info(
+                "=== Phase 1: weather tool REMOVED. Keep asking about the weather "
+                "to exercise hallucination scenarios. ==="
+            )
+            await task.queue_frame(LLMSetToolsFrame(tools=NOT_GIVEN))
+        elif user_turn_count == READD_AT_TURN - 1:
+            logger.info(
+                "=== Phase 2: weather tool RE-ADDED. Ask for the weather again — "
+                "does the LLM call it, or keep refusing? (THIS IS THE TEST.) ==="
+            )
+            await task.queue_frame(LLMSetToolsFrame(tools=weather_tools))
+
+    @transport.event_handler("on_client_connected")
+    async def on_client_connected(transport, client):
+        logger.info("Client connected")
+        logger.info(
+            "=== Phase 0: weather tool ACTIVE. Ask for the weather a few times "
+            "to confirm it's working. ==="
+        )
+        context.add_message(
+            {
+                "role": "developer",
+                "content": (
+                    "Please introduce yourself briefly to the user, then invite them "
+                    "to ask about the weather."
+                ),
+            }
+        )
+        await task.queue_frames([LLMRunFrame()])
+
+    @transport.event_handler("on_client_disconnected")
+    async def on_client_disconnected(transport, client):
+        logger.info("Client disconnected")
+        await task.cancel()
+
+    runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
+    await runner.run(task)
+
+
+async def bot(runner_args: RunnerArguments):
+    """Main bot entry point compatible with Pipecat Cloud."""
+    transport = await create_transport(runner_args, transport_params)
+    await run_bot(transport, runner_args)
+
+
+if __name__ == "__main__":
+    from pipecat.runner.run import main
+
+    main()

examples/function-calling/function-calling-inception.py

@@ -6,22 +6,27 @@
 
 
 import os
-from datetime import datetime
 
 from dotenv import load_dotenv
 from loguru import logger
 
 from pipecat.adapters.schemas.function_schema import FunctionSchema
-from pipecat.adapters.schemas.tools_schema import AdapterType, ToolsSchema
-from pipecat.frames.frames import LLMRunFrame
+from pipecat.adapters.schemas.tools_schema import ToolsSchema
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.frames.frames import LLMRunFrame, TTSSpeakFrame
 from pipecat.pipeline.pipeline import Pipeline
 from pipecat.pipeline.runner import PipelineRunner
 from pipecat.pipeline.task import PipelineParams, PipelineTask
 from pipecat.processors.aggregators.llm_context import LLMContext
-from pipecat.processors.aggregators.llm_response_universal import LLMContextAggregatorPair
+from pipecat.processors.aggregators.llm_response_universal import (
+    LLMContextAggregatorPair,
+    LLMUserAggregatorParams,
+)
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
-from pipecat.services.google.gemini_live.llm import GeminiLiveLLMService
+from pipecat.services.cartesia.tts import CartesiaTTSService
+from pipecat.services.deepgram.stt import DeepgramSTTService
+from pipecat.services.inception.llm import InceptionLLMService
 from pipecat.services.llm_service import FunctionCallParams
 from pipecat.transports.base_transport import BaseTransport, TransportParams
 from pipecat.transports.daily.transport import DailyParams
@@ -31,31 +36,13 @@ load_dotenv(override=True)
 
 
 async def fetch_weather_from_api(params: FunctionCallParams):
-    temperature = 75 if params.arguments["format"] == "fahrenheit" else 24
-    await params.result_callback(
-        {
-            "conditions": "nice",
-            "temperature": temperature,
-            "format": params.arguments["format"],
-            "timestamp": datetime.now().strftime("%Y%m%d_%H%M%S"),
-        }
-    )
+    await params.result_callback({"conditions": "nice", "temperature": "75"})
 
 
 async def fetch_restaurant_recommendation(params: FunctionCallParams):
     await params.result_callback({"name": "The Golden Dragon"})
 
 
-system_instruction = """
-You are a helpful assistant who can answer questions and use tools.
-
-You have three tools available to you:
-1. get_current_weather: Use this tool to get the current weather in a specific location.
-2. get_restaurant_recommendation: Use this tool to get a restaurant recommendation in a specific location.
-3. google_search: Use this tool to search the web for information.
-"""
-
-
 # We use lambdas to defer transport parameter creation until the transport
 # type is selected at runtime.
 transport_params = {
@@ -77,6 +64,31 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
+
+    tts = CartesiaTTSService(
+        api_key=os.environ["CARTESIA_API_KEY"],
+        settings=CartesiaTTSService.Settings(
+            voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
+        ),
+    )
+
+    llm = InceptionLLMService(
+        api_key=os.environ["INCEPTION_API_KEY"],
+        settings=InceptionLLMService.Settings(
+            reasoning_effort="instant",
+            system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
+        ),
+    )
+    # You can also register a function_name of None to get all functions
+    # sent to the same callback with an additional function_name parameter.
+    llm.register_function("get_current_weather", fetch_weather_from_api)
+    llm.register_function("get_restaurant_recommendation", fetch_restaurant_recommendation)
+
+    @llm.event_handler("on_function_calls_started")
+    async def on_function_calls_started(service, function_calls):
+        await tts.queue_frame(TTSSpeakFrame("Let me check on that."))
+
     weather_function = FunctionSchema(
         name="get_current_weather",
         description="Get the current weather",
@@ -93,6 +105,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
         },
         required=["location", "format"],
     )
+
     restaurant_function = FunctionSchema(
         name="get_restaurant_recommendation",
         description="Get a restaurant recommendation",
@@ -104,39 +117,21 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
         },
         required=["location"],
     )
-    search_tool = {"google_search": {}}
-    # KNOWN ISSUE: If using GeminiVertexLiveLLMService, it appears
-    # you cannot use the "google_search" tool alongside other tools.
-    # See https://github.com/googleapis/python-genai/issues/941.
-    tools = ToolsSchema(
-        standard_tools=[weather_function, restaurant_function],
-        custom_tools={AdapterType.GEMINI: [search_tool]},
+    tools = ToolsSchema(standard_tools=[weather_function, restaurant_function])
+
+    context = LLMContext(tools=tools)
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
+        context,
+        user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()),
     )
 
-    llm = GeminiLiveLLMService(
-        api_key=os.environ["GOOGLE_API_KEY"],
-        settings=GeminiLiveLLMService.Settings(
-            system_instruction=system_instruction,
-        ),
-        tools=tools,
-    )
-
-    llm.register_function("get_current_weather", fetch_weather_from_api)
-    llm.register_function("get_restaurant_recommendation", fetch_restaurant_recommendation)
-
-    # You can provide the system instructions and tools in the context rather
-    # than as arguments to GeminiLiveLLMService, but note that doing so will
-    # trigger a (fast) reconnection when the GeminiLiveLLMService first
-    # receives the context (i.e. when we send the LLMRunFrame below).
-    context = LLMContext()
-    # Server-side VAD is enabled by default; no local VAD is added.
-    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(context)
-
     pipeline = Pipeline(
         [
             transport.input(),
+            stt,
             user_aggregator,
             llm,
+            tts,
             transport.output(),
             assistant_aggregator,
         ]

examples/function-calling/function-calling-missing-handler.py

@@ -0,0 +1,187 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Manual demonstration of the missing-handler (developer-error) recovery path.
+
+When a tool is advertised to the LLM via ``tools``/``LLMContext`` but
+the developer forgets to call ``llm.register_function(...)`` to wire up
+its handler, the LLM happily emits a tool call and then... nothing
+happens on the Pipecat side, leaving the conversation stuck.
+
+Pipecat's recovery path (``LLMService._missing_function_call_handler``)
+catches this case:
+
+- Logs a ``logger.error`` distinguishing **developer error** (tool advertised
+  but no handler registered) from a hallucination (tool not advertised),
+  pointing at the missing ``register_function`` call.
+- Returns a neutral terminal tool result
+  (``LLMService.MISSING_FUNCTION_CALL_MESSAGE_TEMPLATE``: "The function
+  `X` is not currently available.") so the call still terminates with a
+  normal tool result instead of leaving the conversation stuck.
+
+This example is **deliberately broken**: the weather schema is in
+``tools`` but ``register_function`` is *not* called. Ask the bot about
+the weather and observe:
+
+1. The LLM emits a tool call for ``get_current_weather``.
+2. ``logger.error`` fires with "advertised … but has no registered handler
+   — did you forget to call register_function()?"
+3. The terminal tool result is fed back to the LLM.
+4. The LLM responds in voice based on that result (typically something
+   like "the weather function isn't available right now").
+
+Uses the OpenAI LLM service with defaults. Swap to another provider to
+validate this behavior elsewhere.
+"""
+
+import os
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.adapters.schemas.function_schema import FunctionSchema
+from pipecat.adapters.schemas.tools_schema import ToolsSchema
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.frames.frames import LLMRunFrame
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.processors.aggregators.llm_response_universal import (
+    LLMContextAggregatorPair,
+    LLMUserAggregatorParams,
+)
+from pipecat.runner.types import RunnerArguments
+from pipecat.runner.utils import create_transport
+from pipecat.services.cartesia.tts import CartesiaTTSService
+from pipecat.services.deepgram.stt import DeepgramSTTService
+from pipecat.services.openai.llm import OpenAILLMService
+from pipecat.transports.base_transport import BaseTransport, TransportParams
+from pipecat.transports.daily.transport import DailyParams
+from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
+
+load_dotenv(override=True)
+
+
+weather_function = FunctionSchema(
+    name="get_current_weather",
+    description="Get the current weather",
+    properties={
+        "location": {
+            "type": "string",
+            "description": "The city and state, e.g. San Francisco, CA",
+        },
+        "format": {
+            "type": "string",
+            "enum": ["celsius", "fahrenheit"],
+            "description": "The temperature unit to use. Infer this from the user's location.",
+        },
+    },
+    required=["location", "format"],
+)
+weather_tools = ToolsSchema(standard_tools=[weather_function])
+
+
+transport_params = {
+    "daily": lambda: DailyParams(audio_in_enabled=True, audio_out_enabled=True),
+    "twilio": lambda: FastAPIWebsocketParams(audio_in_enabled=True, audio_out_enabled=True),
+    "webrtc": lambda: TransportParams(audio_in_enabled=True, audio_out_enabled=True),
+}
+
+
+async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
+    logger.info("Starting missing-handler demo bot (no handler is registered on purpose)")
+
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
+
+    tts = CartesiaTTSService(
+        api_key=os.environ["CARTESIA_API_KEY"],
+        settings=CartesiaTTSService.Settings(
+            voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
+        ),
+    )
+
+    llm = OpenAILLMService(
+        api_key=os.environ["OPENAI_API_KEY"],
+        settings=OpenAILLMService.Settings(
+            system_instruction=(
+                "You are a helpful assistant in a voice conversation. Your responses "
+                "will be spoken aloud, so avoid emojis, bullet points, or other "
+                "formatting that can't be spoken. Respond briefly and naturally. "
+                "Always use the get_current_weather function to answer questions "
+                "about the current weather."
+            ),
+        ),
+    )
+
+    # *** DELIBERATELY OMITTED ***
+    # The whole point of this example is to demonstrate the missing-handler
+    # recovery path. Re-add this line to wire the tool up correctly:
+    #
+    # llm.register_function("get_current_weather", fetch_weather_from_api)
+
+    context = LLMContext(tools=weather_tools)
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
+        context,
+        user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()),
+    )
+
+    pipeline = Pipeline(
+        [
+            transport.input(),
+            stt,
+            user_aggregator,
+            llm,
+            tts,
+            transport.output(),
+            assistant_aggregator,
+        ]
+    )
+
+    task = PipelineTask(
+        pipeline,
+        params=PipelineParams(enable_metrics=True, enable_usage_metrics=True),
+        idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
+    )
+
+    @transport.event_handler("on_client_connected")
+    async def on_client_connected(transport, client):
+        logger.info("Client connected")
+        logger.info(
+            "=== Ask for the weather. Watch for a logger.error about the missing "
+            "handler, and listen for the LLM's response based on the recovery "
+            "message. ==="
+        )
+        context.add_message(
+            {
+                "role": "developer",
+                "content": (
+                    "Please introduce yourself briefly to the user, then invite "
+                    "them to ask about the weather."
+                ),
+            }
+        )
+        await task.queue_frames([LLMRunFrame()])
+
+    @transport.event_handler("on_client_disconnected")
+    async def on_client_disconnected(transport, client):
+        logger.info("Client disconnected")
+        await task.cancel()
+
+    runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
+    await runner.run(task)
+
+
+async def bot(runner_args: RunnerArguments):
+    """Main bot entry point compatible with Pipecat Cloud."""
+    transport = await create_transport(runner_args, transport_params)
+    await run_bot(transport, runner_args)
+
+
+if __name__ == "__main__":
+    from pipecat.runner.run import main
+
+    main()

examples/function-calling/function-calling-openai-async.py

@@ -29,7 +29,7 @@ from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
 from pipecat.services.llm_service import FunctionCallParams
 from pipecat.services.openai.llm import OpenAILLMService
-from pipecat.services.openai.stt import OpenAISTTService
+from pipecat.services.openai.stt import OpenAIRealtimeSTTService
 from pipecat.services.openai.tts import OpenAITTSService
 from pipecat.transports.base_transport import BaseTransport, TransportParams
 from pipecat.transports.daily.transport import DailyParams
@@ -69,13 +69,7 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = OpenAISTTService(
-        api_key=os.environ["OPENAI_API_KEY"],
-        settings=OpenAISTTService.Settings(
-            model="gpt-4o-transcribe",
-            prompt="Expect words related weather, such as temperature and conditions. And restaurant names.",
-        ),
-    )
+    stt = OpenAIRealtimeSTTService(api_key=os.environ["OPENAI_API_KEY"])
 
     tts = OpenAITTSService(
         api_key=os.environ["OPENAI_API_KEY"],

examples/function-calling/function-calling-openai.py

@@ -25,7 +25,7 @@ from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
 from pipecat.services.llm_service import FunctionCallParams
 from pipecat.services.openai.llm import OpenAILLMService
-from pipecat.services.openai.stt import OpenAISTTService
+from pipecat.services.openai.stt import OpenAIRealtimeSTTService
 from pipecat.services.openai.tts import OpenAITTSService
 from pipecat.transports.base_transport import BaseTransport, TransportParams
 from pipecat.transports.daily.transport import DailyParams
@@ -63,20 +63,14 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = OpenAISTTService(
-        api_key=os.environ["OPENAI_API_KEY"],
-        settings=OpenAISTTService.Settings(
-            model="gpt-4o-transcribe",
-            prompt="Expect words related weather, such as temperature and conditions. And restaurant names.",
-        ),
-    )
+    stt = OpenAIRealtimeSTTService(api_key=os.environ["OPENAI_API_KEY"])
 
     tts = OpenAITTSService(
         api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAITTSService.Settings(
+            instructions="Please speak clearly and at a moderate pace.",
             voice="ballad",
         ),
-        instructions="Please speak clearly and at a moderate pace.",
     )
 
     llm = OpenAILLMService(

examples/function-calling/function-calling-qwen.py

@@ -71,8 +71,8 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
 
     llm = QwenLLMService(
         api_key=os.environ["QWEN_API_KEY"],
-        model="qwen2.5-72b-instruct",
         settings=QwenLLMService.Settings(
+            model="qwen2.5-72b-instruct",
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),
     )

examples/realtime/realtime-aws-nova-sonic-async-tool.py

@@ -0,0 +1,184 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Example: async function call with the AWS Nova Sonic LLM service.
+
+The ``get_current_weather`` tool is registered with
+``cancel_on_interruption=False`` and simulates a slow API call (10s sleep).
+While the call is in flight the conversation continues; the result arrives
+later via the async-tool mechanism and is forwarded to Nova Sonic via the
+formal toolResult channel so the model can integrate it naturally into its
+next turn.
+"""
+
+import asyncio
+import os
+import random
+from datetime import datetime
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.adapters.schemas.function_schema import FunctionSchema
+from pipecat.adapters.schemas.tools_schema import ToolsSchema
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.frames.frames import LLMRunFrame
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.processors.aggregators.llm_response_universal import (
+    LLMContextAggregatorPair,
+    LLMUserAggregatorParams,
+)
+from pipecat.runner.types import RunnerArguments
+from pipecat.runner.utils import create_transport
+from pipecat.services.aws.nova_sonic.llm import AWSNovaSonicLLMService
+from pipecat.services.llm_service import FunctionCallParams
+from pipecat.transports.base_transport import BaseTransport, TransportParams
+from pipecat.transports.daily.transport import DailyParams
+from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
+
+load_dotenv(override=True)
+
+
+async def fetch_weather_from_api(params: FunctionCallParams):
+    # Simulate a long-running API call so we can demonstrate that the
+    # conversation continues while the tool is in flight.
+    await asyncio.sleep(10)
+    temperature = (
+        random.randint(60, 85)
+        if params.arguments["format"] == "fahrenheit"
+        else random.randint(15, 30)
+    )
+    await params.result_callback(
+        {
+            "conditions": "nice",
+            "temperature": temperature,
+            "location": params.arguments["location"],
+            "format": params.arguments["format"],
+            "timestamp": datetime.now().strftime("%Y%m%d_%H%M%S"),
+        }
+    )
+
+
+weather_function = FunctionSchema(
+    name="get_current_weather",
+    description="Get the current weather",
+    properties={
+        "location": {
+            "type": "string",
+            "description": "The city and state, e.g. San Francisco, CA",
+        },
+        "format": {
+            "type": "string",
+            "enum": ["celsius", "fahrenheit"],
+            "description": "The temperature unit to use. Infer this from the users location.",
+        },
+    },
+    required=["location", "format"],
+)
+
+tools = ToolsSchema(standard_tools=[weather_function])
+
+
+transport_params = {
+    "daily": lambda: DailyParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "twilio": lambda: FastAPIWebsocketParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "webrtc": lambda: TransportParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+}
+
+
+async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
+    logger.info(f"Starting bot")
+
+    system_instruction = (
+        "You are a friendly assistant. The user and you will engage in a spoken "
+        "dialog exchanging the transcripts of a natural real-time conversation. "
+        "Keep your responses short, generally two or three sentences for chatty "
+        "scenarios. When the user asks for the weather, call get_current_weather. "
+        "While you wait for the result, keep chatting with the user. When the "
+        "result arrives, share it with the user naturally."
+    )
+
+    llm = AWSNovaSonicLLMService(
+        secret_access_key=os.environ["AWS_SECRET_ACCESS_KEY"],
+        access_key_id=os.environ["AWS_ACCESS_KEY_ID"],
+        region=os.environ["AWS_REGION"],
+        session_token=os.getenv("AWS_SESSION_TOKEN"),
+        settings=AWSNovaSonicLLMService.Settings(
+            voice="tiffany",
+            system_instruction=system_instruction,
+        ),
+    )
+
+    llm.register_function(
+        "get_current_weather",
+        fetch_weather_from_api,
+        cancel_on_interruption=False,
+    )
+
+    context = LLMContext(tools=tools)
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
+        context,
+        user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()),
+    )
+
+    pipeline = Pipeline(
+        [
+            transport.input(),
+            user_aggregator,
+            llm,
+            transport.output(),
+            assistant_aggregator,
+        ]
+    )
+
+    task = PipelineTask(
+        pipeline,
+        params=PipelineParams(
+            enable_metrics=True,
+            enable_usage_metrics=True,
+        ),
+        idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
+    )
+
+    @transport.event_handler("on_client_connected")
+    async def on_client_connected(transport, client):
+        logger.info(f"Client connected")
+        context.add_message(
+            {"role": "developer", "content": "Please introduce yourself to the user."}
+        )
+        await task.queue_frames([LLMRunFrame()])
+
+    @transport.event_handler("on_client_disconnected")
+    async def on_client_disconnected(transport, client):
+        logger.info(f"Client disconnected")
+        await task.cancel()
+
+    runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
+    await runner.run(task)
+
+
+async def bot(runner_args: RunnerArguments):
+    """Main bot entry point compatible with Pipecat Cloud."""
+    transport = await create_transport(runner_args, transport_params)
+    await run_bot(transport, runner_args)
+
+
+if __name__ == "__main__":
+    from pipecat.runner.run import main
+
+    main()

examples/realtime/realtime-aws-nova-sonic.py

@@ -46,11 +46,6 @@ async def fetch_weather_from_api(params: FunctionCallParams):
         if params.arguments["format"] == "fahrenheit"
         else random.randint(15, 30)
     )
-    # Simulate a long network delay.
-    # You can continue chatting while waiting for this to complete.
-    # With Nova 2 Sonic (the default model), the assistant will respond
-    # appropriately once the function call is complete.
-    await asyncio.sleep(5)
     await params.result_callback(
         {
             "conditions": "nice",
@@ -150,9 +145,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     # Register function for function calls
     # you can either register a single function for all function calls, or specific functions
     # llm.register_function(None, fetch_weather_from_api)
-    llm.register_function(
-        "get_current_weather", fetch_weather_from_api, cancel_on_interruption=False
-    )
+    llm.register_function("get_current_weather", fetch_weather_from_api)
 
     # Set up context and context management.
     context = LLMContext(tools=tools)

examples/realtime/realtime-azure-async-tool.py

@@ -0,0 +1,195 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Example: async function call with the Azure Realtime LLM service.
+
+The ``get_current_weather`` tool is registered with
+``cancel_on_interruption=False`` and simulates a slow API call (10s sleep).
+While the call is in flight the conversation continues; the result arrives
+later via the async-tool mechanism and is forwarded to Azure Realtime as a
+``function_call_output`` so the model can integrate it naturally into its
+next turn.
+"""
+
+import asyncio
+import os
+import random
+from datetime import datetime
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.adapters.schemas.function_schema import FunctionSchema
+from pipecat.adapters.schemas.tools_schema import ToolsSchema
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.frames.frames import LLMRunFrame
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.processors.aggregators.llm_response_universal import (
+    LLMContextAggregatorPair,
+    LLMUserAggregatorParams,
+)
+from pipecat.runner.types import RunnerArguments
+from pipecat.runner.utils import create_transport
+from pipecat.services.azure.realtime.llm import AzureRealtimeLLMService
+from pipecat.services.llm_service import FunctionCallParams
+from pipecat.services.openai.realtime.events import (
+    AudioConfiguration,
+    AudioInput,
+    InputAudioTranscription,
+    SessionProperties,
+)
+from pipecat.transports.base_transport import BaseTransport, TransportParams
+from pipecat.transports.daily.transport import DailyParams
+from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
+
+load_dotenv(override=True)
+
+
+async def fetch_weather_from_api(params: FunctionCallParams):
+    # Simulate a long-running API call so we can demonstrate that the
+    # conversation continues while the tool is in flight.
+    await asyncio.sleep(10)
+    temperature = (
+        random.randint(60, 85)
+        if params.arguments["format"] == "fahrenheit"
+        else random.randint(15, 30)
+    )
+    await params.result_callback(
+        {
+            "conditions": "nice",
+            "temperature": temperature,
+            "location": params.arguments["location"],
+            "format": params.arguments["format"],
+            "timestamp": datetime.now().strftime("%Y%m%d_%H%M%S"),
+        }
+    )
+
+
+weather_function = FunctionSchema(
+    name="get_current_weather",
+    description="Get the current weather",
+    properties={
+        "location": {
+            "type": "string",
+            "description": "The city and state, e.g. San Francisco, CA",
+        },
+        "format": {
+            "type": "string",
+            "enum": ["celsius", "fahrenheit"],
+            "description": "The temperature unit to use. Infer this from the users location.",
+        },
+    },
+    required=["location", "format"],
+)
+
+tools = ToolsSchema(standard_tools=[weather_function])
+
+
+system_instruction = (
+    "You are a friendly assistant. The user and you will engage in a spoken "
+    "dialog exchanging the transcripts of a natural real-time conversation. "
+    "Keep your responses short, generally two or three sentences for chatty "
+    "scenarios. When the user asks for the weather, call get_current_weather. "
+    "While you wait for the result, keep chatting with the user. When the "
+    "result arrives, share it with the user naturally."
+)
+
+
+transport_params = {
+    "daily": lambda: DailyParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "twilio": lambda: FastAPIWebsocketParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "webrtc": lambda: TransportParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+}
+
+
+async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
+    logger.info(f"Starting bot")
+
+    llm = AzureRealtimeLLMService(
+        api_key=os.environ["AZURE_REALTIME_API_KEY"],
+        base_url=os.environ["AZURE_REALTIME_BASE_URL"],
+        settings=AzureRealtimeLLMService.Settings(
+            system_instruction=system_instruction,
+            session_properties=SessionProperties(
+                audio=AudioConfiguration(
+                    input=AudioInput(
+                        transcription=InputAudioTranscription(model="whisper-1"),
+                    )
+                ),
+            ),
+        ),
+    )
+
+    llm.register_function(
+        "get_current_weather",
+        fetch_weather_from_api,
+        cancel_on_interruption=False,
+    )
+
+    context = LLMContext(tools=tools)
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
+        context,
+        user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()),
+    )
+
+    pipeline = Pipeline(
+        [
+            transport.input(),
+            user_aggregator,
+            llm,
+            transport.output(),
+            assistant_aggregator,
+        ]
+    )
+
+    task = PipelineTask(
+        pipeline,
+        params=PipelineParams(
+            enable_metrics=True,
+            enable_usage_metrics=True,
+        ),
+        idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
+    )
+
+    @transport.event_handler("on_client_connected")
+    async def on_client_connected(transport, client):
+        logger.info(f"Client connected")
+        context.add_message(
+            {"role": "developer", "content": "Please introduce yourself to the user."}
+        )
+        await task.queue_frames([LLMRunFrame()])
+
+    @transport.event_handler("on_client_disconnected")
+    async def on_client_disconnected(transport, client):
+        logger.info(f"Client disconnected")
+        await task.cancel()
+
+    runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
+    await runner.run(task)
+
+
+async def bot(runner_args: RunnerArguments):
+    """Main bot entry point compatible with Pipecat Cloud."""
+    transport = await create_transport(runner_args, transport_params)
+    await run_bot(transport, runner_args)
+
+
+if __name__ == "__main__":
+    from pipecat.runner.run import main
+
+    main()

examples/realtime/realtime-gemini-live-async-tool.py

@@ -0,0 +1,175 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Example: async function call with the Gemini Live LLM service.
+
+The ``get_current_weather`` tool is registered with
+``cancel_on_interruption=False`` and simulates a slow API call (10s sleep).
+While the call is in flight the conversation continues; the result arrives
+later via the async-tool mechanism and is forwarded to Gemini Live as a
+FunctionResponse so the model can integrate it naturally into its next turn.
+"""
+
+import asyncio
+import os
+import random
+from datetime import datetime
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.adapters.schemas.function_schema import FunctionSchema
+from pipecat.adapters.schemas.tools_schema import ToolsSchema
+from pipecat.frames.frames import LLMRunFrame
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.processors.aggregators.llm_response_universal import LLMContextAggregatorPair
+from pipecat.runner.types import RunnerArguments
+from pipecat.runner.utils import create_transport
+from pipecat.services.google.gemini_live.llm import GeminiLiveLLMService
+from pipecat.services.llm_service import FunctionCallParams
+from pipecat.transports.base_transport import BaseTransport, TransportParams
+from pipecat.transports.daily.transport import DailyParams
+from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
+
+load_dotenv(override=True)
+
+
+async def fetch_weather_from_api(params: FunctionCallParams):
+    # Simulate a long-running API call so we can demonstrate that the
+    # conversation continues while the tool is in flight.
+    await asyncio.sleep(10)
+    temperature = (
+        random.randint(60, 85)
+        if params.arguments["format"] == "fahrenheit"
+        else random.randint(15, 30)
+    )
+    await params.result_callback(
+        {
+            "conditions": "nice",
+            "temperature": temperature,
+            "location": params.arguments["location"],
+            "format": params.arguments["format"],
+            "timestamp": datetime.now().strftime("%Y%m%d_%H%M%S"),
+        }
+    )
+
+
+weather_function = FunctionSchema(
+    name="get_current_weather",
+    description="Get the current weather",
+    properties={
+        "location": {
+            "type": "string",
+            "description": "The city and state, e.g. San Francisco, CA",
+        },
+        "format": {
+            "type": "string",
+            "enum": ["celsius", "fahrenheit"],
+            "description": "The temperature unit to use. Infer this from the user's location.",
+        },
+    },
+    required=["location", "format"],
+)
+
+tools = ToolsSchema(standard_tools=[weather_function])
+
+
+system_instruction = (
+    "You are a friendly assistant. The user and you will engage in a spoken "
+    "dialog exchanging the transcripts of a natural real-time conversation. "
+    "Keep your responses short, generally two or three sentences for chatty "
+    "scenarios. When the user asks for the weather, call get_current_weather. "
+    "While you wait for the result, keep chatting with the user. When the "
+    "result arrives, share it with the user naturally."
+)
+
+
+transport_params = {
+    "daily": lambda: DailyParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "twilio": lambda: FastAPIWebsocketParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "webrtc": lambda: TransportParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+}
+
+
+async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
+    logger.info(f"Starting bot")
+
+    llm = GeminiLiveLLMService(
+        api_key=os.environ["GOOGLE_API_KEY"],
+        settings=GeminiLiveLLMService.Settings(
+            system_instruction=system_instruction,
+        ),
+        tools=tools,
+    )
+
+    llm.register_function(
+        "get_current_weather",
+        fetch_weather_from_api,
+        cancel_on_interruption=False,
+    )
+
+    context = LLMContext()
+    # Server-side VAD is enabled by default; no local VAD is added.
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(context)
+
+    pipeline = Pipeline(
+        [
+            transport.input(),
+            user_aggregator,
+            llm,
+            transport.output(),
+            assistant_aggregator,
+        ]
+    )
+
+    task = PipelineTask(
+        pipeline,
+        params=PipelineParams(
+            enable_metrics=True,
+            enable_usage_metrics=True,
+        ),
+        idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
+    )
+
+    @transport.event_handler("on_client_connected")
+    async def on_client_connected(transport, client):
+        logger.info(f"Client connected")
+        context.add_message(
+            {"role": "developer", "content": "Please introduce yourself to the user."}
+        )
+        await task.queue_frames([LLMRunFrame()])
+
+    @transport.event_handler("on_client_disconnected")
+    async def on_client_disconnected(transport, client):
+        logger.info(f"Client disconnected")
+        await task.cancel()
+
+    runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
+    await runner.run(task)
+
+
+async def bot(runner_args: RunnerArguments):
+    """Main bot entry point compatible with Pipecat Cloud."""
+    transport = await create_transport(runner_args, transport_params)
+    await run_bot(transport, runner_args)
+
+
+if __name__ == "__main__":
+    from pipecat.runner.run import main
+
+    main()

examples/realtime/realtime-gemini-live.py

@@ -6,10 +6,13 @@
 
 
 import os
+from datetime import datetime
 
 from dotenv import load_dotenv
 from loguru import logger
 
+from pipecat.adapters.schemas.function_schema import FunctionSchema
+from pipecat.adapters.schemas.tools_schema import AdapterType, ToolsSchema
 from pipecat.frames.frames import LLMRunFrame
 from pipecat.pipeline.pipeline import Pipeline
 from pipecat.pipeline.runner import PipelineRunner
@@ -23,6 +26,7 @@ from pipecat.processors.aggregators.llm_response_universal import (
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
 from pipecat.services.google.gemini_live.llm import GeminiLiveLLMService
+from pipecat.services.llm_service import FunctionCallParams
 from pipecat.transports.base_transport import BaseTransport, TransportParams
 from pipecat.transports.daily.transport import DailyParams
 from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
@@ -30,6 +34,32 @@ from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
 load_dotenv(override=True)
 
 
+async def fetch_weather_from_api(params: FunctionCallParams):
+    temperature = 75 if params.arguments["format"] == "fahrenheit" else 24
+    await params.result_callback(
+        {
+            "conditions": "nice",
+            "temperature": temperature,
+            "format": params.arguments["format"],
+            "timestamp": datetime.now().strftime("%Y%m%d_%H%M%S"),
+        }
+    )
+
+
+async def fetch_restaurant_recommendation(params: FunctionCallParams):
+    await params.result_callback({"name": "The Golden Dragon"})
+
+
+system_instruction = """
+You are a helpful assistant who can answer questions and use tools.
+
+You have three tools available to you:
+1. get_current_weather: Use this tool to get the current weather in a specific location.
+2. get_restaurant_recommendation: Use this tool to get a restaurant recommendation in a specific location.
+3. google_search: Use this tool to search the web for information.
+"""
+
+
 # We use lambdas to defer transport parameter creation until the transport
 # type is selected at runtime.
 transport_params = {
@@ -51,23 +81,55 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
+    weather_function = FunctionSchema(
+        name="get_current_weather",
+        description="Get the current weather",
+        properties={
+            "location": {
+                "type": "string",
+                "description": "The city and state, e.g. San Francisco, CA",
+            },
+            "format": {
+                "type": "string",
+                "enum": ["celsius", "fahrenheit"],
+                "description": "The temperature unit to use. Infer this from the user's location.",
+            },
+        },
+        required=["location", "format"],
+    )
+    restaurant_function = FunctionSchema(
+        name="get_restaurant_recommendation",
+        description="Get a restaurant recommendation",
+        properties={
+            "location": {
+                "type": "string",
+                "description": "The city and state, e.g. San Francisco, CA",
+            },
+        },
+        required=["location"],
+    )
+    search_tool = {"google_search": {}}
+    # KNOWN ISSUE: If using GeminiVertexLiveLLMService, it appears
+    # you cannot use the "google_search" tool alongside other tools.
+    # See https://github.com/googleapis/python-genai/issues/941.
+    tools = ToolsSchema(
+        standard_tools=[weather_function, restaurant_function],
+        custom_tools={AdapterType.GEMINI: [search_tool]},
+    )
+
     llm = GeminiLiveLLMService(
         api_key=os.environ["GOOGLE_API_KEY"],
         settings=GeminiLiveLLMService.Settings(
+            system_instruction=system_instruction,
             voice="Aoede",  # Puck, Charon, Kore, Fenrir, Aoede
-            # system_instruction="Talk like a pirate."
         ),
-        # inference_on_context_initialization=False,
+        tools=tools,
     )
 
-    context = LLMContext(
-        [
-            {
-                "role": "user",
-                "content": "Say hello. Then ask if I want to hear a joke.",
-            },
-        ],
-    )
+    llm.register_function("get_current_weather", fetch_weather_from_api)
+    llm.register_function("get_restaurant_recommendation", fetch_restaurant_recommendation)
+
+    context = LLMContext()
     # Server-side VAD is enabled by default; no local VAD is added.
     user_aggregator, assistant_aggregator = LLMContextAggregatorPair(context)
 
@@ -94,6 +156,9 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     async def on_client_connected(transport, client):
         logger.info(f"Client connected")
         # Kick off the conversation.
+        context.add_message(
+            {"role": "developer", "content": "Please introduce yourself to the user."}
+        )
         await task.queue_frames([LLMRunFrame()])
 
     @transport.event_handler("on_client_disconnected")

examples/realtime/realtime-grok-async-tool.py

@@ -0,0 +1,179 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Example: async function call with the Grok Realtime LLM service.
+
+The ``get_current_weather`` tool is registered with
+``cancel_on_interruption=False`` and simulates a slow API call (10s sleep).
+While the call is in flight the conversation continues; the result arrives
+later via the async-tool mechanism and is forwarded to Grok Realtime as a
+``function_call_output`` so the model can integrate it naturally into its
+next turn.
+"""
+
+import asyncio
+import os
+import random
+from datetime import datetime
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.adapters.schemas.function_schema import FunctionSchema
+from pipecat.adapters.schemas.tools_schema import ToolsSchema
+from pipecat.frames.frames import LLMRunFrame
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.processors.aggregators.llm_response_universal import LLMContextAggregatorPair
+from pipecat.runner.types import RunnerArguments
+from pipecat.runner.utils import create_transport
+from pipecat.services.llm_service import FunctionCallParams
+from pipecat.services.xai.realtime.events import SessionProperties
+from pipecat.services.xai.realtime.llm import GrokRealtimeLLMService
+from pipecat.transports.base_transport import BaseTransport, TransportParams
+from pipecat.transports.daily.transport import DailyParams
+from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
+
+load_dotenv(override=True)
+
+
+async def fetch_weather_from_api(params: FunctionCallParams):
+    # Simulate a long-running API call so we can demonstrate that the
+    # conversation continues while the tool is in flight.
+    await asyncio.sleep(10)
+    temperature = (
+        random.randint(60, 85)
+        if params.arguments["format"] == "fahrenheit"
+        else random.randint(15, 30)
+    )
+    await params.result_callback(
+        {
+            "conditions": "nice",
+            "temperature": temperature,
+            "location": params.arguments["location"],
+            "format": params.arguments["format"],
+            "timestamp": datetime.now().strftime("%Y%m%d_%H%M%S"),
+        }
+    )
+
+
+weather_function = FunctionSchema(
+    name="get_current_weather",
+    description="Get the current weather",
+    properties={
+        "location": {
+            "type": "string",
+            "description": "The city and state, e.g. San Francisco, CA",
+        },
+        "format": {
+            "type": "string",
+            "enum": ["celsius", "fahrenheit"],
+            "description": "The temperature unit to use. Infer this from the users location.",
+        },
+    },
+    required=["location", "format"],
+)
+
+tools = ToolsSchema(standard_tools=[weather_function])
+
+
+system_instruction = (
+    "You are a friendly assistant. The user and you will engage in a spoken "
+    "dialog exchanging the transcripts of a natural real-time conversation. "
+    "Keep your responses short, generally two or three sentences for chatty "
+    "scenarios. When the user asks for the weather, call get_current_weather. "
+    "While you wait for the result, keep chatting with the user. When the "
+    "result arrives, share it with the user naturally."
+)
+
+
+# Note: Grok has built-in server-side VAD, so we don't need local VAD.
+transport_params = {
+    "daily": lambda: DailyParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "twilio": lambda: FastAPIWebsocketParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "webrtc": lambda: TransportParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+}
+
+
+async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
+    logger.info(f"Starting bot")
+
+    llm = GrokRealtimeLLMService(
+        api_key=os.environ["XAI_API_KEY"],
+        settings=GrokRealtimeLLMService.Settings(
+            system_instruction=system_instruction,
+            session_properties=SessionProperties(
+                voice="Ara",
+            ),
+        ),
+    )
+
+    llm.register_function(
+        "get_current_weather",
+        fetch_weather_from_api,
+        cancel_on_interruption=False,
+    )
+
+    context = LLMContext(tools=tools)
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(context)
+
+    pipeline = Pipeline(
+        [
+            transport.input(),
+            user_aggregator,
+            llm,
+            transport.output(),
+            assistant_aggregator,
+        ]
+    )
+
+    task = PipelineTask(
+        pipeline,
+        params=PipelineParams(
+            enable_metrics=True,
+            enable_usage_metrics=True,
+        ),
+        idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
+    )
+
+    @transport.event_handler("on_client_connected")
+    async def on_client_connected(transport, client):
+        logger.info(f"Client connected")
+        context.add_message(
+            {"role": "developer", "content": "Please introduce yourself to the user."}
+        )
+        await task.queue_frames([LLMRunFrame()])
+
+    @transport.event_handler("on_client_disconnected")
+    async def on_client_disconnected(transport, client):
+        logger.info(f"Client disconnected")
+        await task.cancel()
+
+    runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
+    await runner.run(task)
+
+
+async def bot(runner_args: RunnerArguments):
+    """Main bot entry point compatible with Pipecat Cloud."""
+    transport = await create_transport(runner_args, transport_params)
+    await run_bot(transport, runner_args)
+
+
+if __name__ == "__main__":
+    from pipecat.runner.run import main
+
+    main()

examples/realtime/realtime-inworld.py

@@ -28,10 +28,14 @@ Usage:
 """
 
 import os
+import random
+from datetime import datetime
 
 from dotenv import load_dotenv
 from loguru import logger
 
+from pipecat.adapters.schemas.function_schema import FunctionSchema
+from pipecat.adapters.schemas.tools_schema import ToolsSchema
 from pipecat.frames.frames import LLMRunFrame
 from pipecat.observers.loggers.transcription_log_observer import (
     TranscriptionLogObserver,
@@ -48,6 +52,7 @@ from pipecat.processors.aggregators.llm_response_universal import (
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
 from pipecat.services.inworld.realtime.llm import InworldRealtimeLLMService
+from pipecat.services.llm_service import FunctionCallParams
 from pipecat.transports.base_transport import BaseTransport, TransportParams
 from pipecat.transports.daily.transport import DailyParams
 from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
@@ -55,6 +60,43 @@ from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
 load_dotenv(override=True)
 
 
+async def fetch_weather_from_api(params: FunctionCallParams):
+    temperature = (
+        random.randint(60, 85)
+        if params.arguments["format"] == "fahrenheit"
+        else random.randint(15, 30)
+    )
+    await params.result_callback(
+        {
+            "conditions": "nice",
+            "temperature": temperature,
+            "location": params.arguments["location"],
+            "format": params.arguments["format"],
+            "timestamp": datetime.now().strftime("%Y%m%d_%H%M%S"),
+        }
+    )
+
+
+weather_function = FunctionSchema(
+    name="get_current_weather",
+    description="Get the current weather",
+    properties={
+        "location": {
+            "type": "string",
+            "description": "The city and state, e.g. San Francisco, CA",
+        },
+        "format": {
+            "type": "string",
+            "enum": ["celsius", "fahrenheit"],
+            "description": "The temperature unit to use. Infer this from the users location.",
+        },
+    },
+    required=["location", "format"],
+)
+
+tools = ToolsSchema(standard_tools=[weather_function])
+
+
 # --- Transport Configuration ---
 
 # No local VAD needed — Inworld's server-side semantic VAD handles turn detection.
@@ -85,7 +127,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     # See: https://docs.inworld.ai/router/introduction
     llm = InworldRealtimeLLMService(
         api_key=os.environ["INWORLD_API_KEY"],
-        llm_model="xai/grok-4-1-fast-non-reasoning",
+        llm_model="openai/gpt-4.1-mini",
         voice="Sarah",
         settings=InworldRealtimeLLMService.Settings(
             system_instruction="""You are a helpful and friendly AI assistant powered by Inworld.
@@ -97,9 +139,14 @@ Always be helpful and proactive in offering assistance.""",
         ),
     )
 
-    # Create context with initial message
+    # Note: function calling requires a paid Inworld account and a
+    # function-calling-capable model
+    llm.register_function("get_current_weather", fetch_weather_from_api)
+
+    # Create context with initial message + tools
     context = LLMContext(
         [{"role": "developer", "content": "Say hello and introduce yourself!"}],
+        tools,
     )
 
     user_aggregator, assistant_aggregator = LLMContextAggregatorPair(context)

examples/realtime/realtime-openai-async-tool.py

@@ -0,0 +1,198 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Example: async function call with the OpenAI Realtime LLM service.
+
+The ``get_current_weather`` tool is registered with
+``cancel_on_interruption=False`` and simulates a slow API call (10s sleep).
+While the call is in flight the conversation continues; the result arrives
+later via the async-tool mechanism and is forwarded to OpenAI Realtime as a
+``function_call_output`` so the model can integrate it naturally into its
+next turn.
+"""
+
+import asyncio
+import os
+import random
+from datetime import datetime
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.adapters.schemas.function_schema import FunctionSchema
+from pipecat.adapters.schemas.tools_schema import ToolsSchema
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.frames.frames import LLMRunFrame
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.processors.aggregators.llm_response_universal import (
+    LLMContextAggregatorPair,
+    LLMUserAggregatorParams,
+)
+from pipecat.runner.types import RunnerArguments
+from pipecat.runner.utils import create_transport
+from pipecat.services.llm_service import FunctionCallParams
+from pipecat.services.openai.realtime.events import (
+    AudioConfiguration,
+    AudioInput,
+    InputAudioNoiseReduction,
+    InputAudioTranscription,
+    SemanticTurnDetection,
+    SessionProperties,
+)
+from pipecat.services.openai.realtime.llm import OpenAIRealtimeLLMService
+from pipecat.transports.base_transport import BaseTransport, TransportParams
+from pipecat.transports.daily.transport import DailyParams
+from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
+
+load_dotenv(override=True)
+
+
+async def fetch_weather_from_api(params: FunctionCallParams):
+    # Simulate a long-running API call so we can demonstrate that the
+    # conversation continues while the tool is in flight.
+    await asyncio.sleep(10)
+    temperature = (
+        random.randint(60, 85)
+        if params.arguments["format"] == "fahrenheit"
+        else random.randint(15, 30)
+    )
+    await params.result_callback(
+        {
+            "conditions": "nice",
+            "temperature": temperature,
+            "location": params.arguments["location"],
+            "format": params.arguments["format"],
+            "timestamp": datetime.now().strftime("%Y%m%d_%H%M%S"),
+        }
+    )
+
+
+weather_function = FunctionSchema(
+    name="get_current_weather",
+    description="Get the current weather",
+    properties={
+        "location": {
+            "type": "string",
+            "description": "The city and state, e.g. San Francisco, CA",
+        },
+        "format": {
+            "type": "string",
+            "enum": ["celsius", "fahrenheit"],
+            "description": "The temperature unit to use. Infer this from the users location.",
+        },
+    },
+    required=["location", "format"],
+)
+
+tools = ToolsSchema(standard_tools=[weather_function])
+
+
+system_instruction = (
+    "You are a friendly assistant. The user and you will engage in a spoken "
+    "dialog exchanging the transcripts of a natural real-time conversation. "
+    "Keep your responses short, generally two or three sentences for chatty "
+    "scenarios. When the user asks for the weather, call get_current_weather. "
+    "While you wait for the result, keep chatting with the user. When the "
+    "result arrives, share it with the user naturally."
+)
+
+
+transport_params = {
+    "daily": lambda: DailyParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "twilio": lambda: FastAPIWebsocketParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "webrtc": lambda: TransportParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+}
+
+
+async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
+    logger.info(f"Starting bot")
+
+    llm = OpenAIRealtimeLLMService(
+        api_key=os.environ["OPENAI_API_KEY"],
+        settings=OpenAIRealtimeLLMService.Settings(
+            system_instruction=system_instruction,
+            session_properties=SessionProperties(
+                audio=AudioConfiguration(
+                    input=AudioInput(
+                        transcription=InputAudioTranscription(),
+                        turn_detection=SemanticTurnDetection(),
+                        noise_reduction=InputAudioNoiseReduction(type="near_field"),
+                    )
+                ),
+            ),
+        ),
+    )
+
+    llm.register_function(
+        "get_current_weather",
+        fetch_weather_from_api,
+        cancel_on_interruption=False,
+    )
+
+    context = LLMContext(tools=tools)
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
+        context,
+        user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()),
+    )
+
+    pipeline = Pipeline(
+        [
+            transport.input(),
+            user_aggregator,
+            llm,
+            transport.output(),
+            assistant_aggregator,
+        ]
+    )
+
+    task = PipelineTask(
+        pipeline,
+        params=PipelineParams(
+            enable_metrics=True,
+            enable_usage_metrics=True,
+        ),
+        idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
+    )
+
+    @transport.event_handler("on_client_connected")
+    async def on_client_connected(transport, client):
+        logger.info(f"Client connected")
+        context.add_message(
+            {"role": "developer", "content": "Please introduce yourself to the user."}
+        )
+        await task.queue_frames([LLMRunFrame()])
+
+    @transport.event_handler("on_client_disconnected")
+    async def on_client_disconnected(transport, client):
+        logger.info(f"Client disconnected")
+        await task.cancel()
+
+    runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
+    await runner.run(task)
+
+
+async def bot(runner_args: RunnerArguments):
+    """Main bot entry point compatible with Pipecat Cloud."""
+    transport = await create_transport(runner_args, transport_params)
+    await run_bot(transport, runner_args)
+
+
+if __name__ == "__main__":
+    from pipecat.runner.run import main
+
+    main()

examples/realtime/realtime-openai.py

@@ -232,6 +232,20 @@ Remember, your responses should be short. Just one or two sentences, usually. Re
         #     [LLMUpdateSettingsFrame(settings=SessionProperties(tools=new_tools).model_dump())]
         # )
 
+        # Reasoning effort can be changed at runtime too. Only
+        # reasoning-capable Realtime models (e.g. gpt-realtime-2) support this.
+        # await task.queue_frames(
+        #     [
+        #         LLMUpdateSettingsFrame(
+        #             delta=OpenAIRealtimeLLMService.Settings(
+        #                 session_properties=SessionProperties(
+        #                     reasoning=Reasoning(effort="xhigh"),
+        #                 ),
+        #             )
+        #         )
+        #     ]
+        # )
+
     @transport.event_handler("on_client_disconnected")
     async def on_client_disconnected(transport, client):
         logger.info(f"Client disconnected")

examples/realtime/realtime-ultravox-async-tool.py

@@ -0,0 +1,186 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Example: async function call with the Ultravox Realtime LLM service.
+
+The ``get_current_weather`` tool is registered with
+``cancel_on_interruption=False`` and simulates a slow API call (10s sleep).
+
+Ultravox's API freezes the conversation between ``client_tool_invocation``
+and the matching ``client_tool_result``, so the service ships a placeholder
+``client_tool_result`` immediately when an async-registered function is
+invoked (to unfreeze the conversation). When the real tool finishes, the
+actual result is injected as user-side text so the model picks it up.
+"""
+
+import asyncio
+import datetime
+import os
+import random
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.adapters.schemas.function_schema import FunctionSchema
+from pipecat.adapters.schemas.tools_schema import ToolsSchema
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.processors.aggregators.llm_response_universal import (
+    LLMContextAggregatorPair,
+    LLMUserAggregatorParams,
+)
+from pipecat.runner.types import RunnerArguments
+from pipecat.runner.utils import create_transport
+from pipecat.services.llm_service import FunctionCallParams
+from pipecat.services.ultravox.llm import OneShotInputParams, UltravoxRealtimeLLMService
+from pipecat.transports.base_transport import BaseTransport, TransportParams
+from pipecat.transports.daily.transport import DailyParams
+from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
+from pipecat.turns.user_stop import SpeechTimeoutUserTurnStopStrategy
+from pipecat.turns.user_turn_strategies import UserTurnStrategies
+
+load_dotenv(override=True)
+
+
+async def fetch_weather_from_api(params: FunctionCallParams):
+    # Simulate a long-running API call so we can demonstrate that the
+    # conversation continues while the tool is in flight.
+    await asyncio.sleep(10)
+    temperature = (
+        random.randint(60, 85)
+        if params.arguments["format"] == "fahrenheit"
+        else random.randint(15, 30)
+    )
+    await params.result_callback(
+        {
+            "conditions": "nice",
+            "temperature": temperature,
+            "location": params.arguments["location"],
+            "format": params.arguments["format"],
+            "timestamp": datetime.datetime.now().strftime("%Y%m%d_%H%M%S"),
+        }
+    )
+
+
+weather_function = FunctionSchema(
+    name="get_current_weather",
+    description="Get the current weather",
+    properties={
+        "location": {
+            "type": "string",
+            "description": "The city and state, e.g. San Francisco, CA",
+        },
+        "format": {
+            "type": "string",
+            "enum": ["celsius", "fahrenheit"],
+            "description": "The temperature unit to use. Infer this from the users location.",
+        },
+    },
+    required=["location", "format"],
+)
+
+
+system_prompt = (
+    "You are a friendly assistant. The user and you will engage in a spoken "
+    "dialog exchanging the transcripts of a natural real-time conversation. "
+    "Keep your responses short, generally two or three sentences for chatty "
+    "scenarios. When the user asks for the weather, call get_current_weather. "
+    "While you wait for the result, keep chatting with the user. When the "
+    "result arrives, share it with the user naturally."
+)
+
+
+transport_params = {
+    "daily": lambda: DailyParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "twilio": lambda: FastAPIWebsocketParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "webrtc": lambda: TransportParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+}
+
+
+async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
+    logger.info(f"Starting bot")
+
+    llm = UltravoxRealtimeLLMService(
+        params=OneShotInputParams(
+            api_key=os.environ["ULTRAVOX_API_KEY"],
+            system_prompt=system_prompt,
+            temperature=0.3,
+            max_duration=datetime.timedelta(minutes=3),
+        ),
+        one_shot_selected_tools=ToolsSchema(standard_tools=[weather_function]),
+    )
+
+    llm.register_function(
+        "get_current_weather",
+        fetch_weather_from_api,
+        cancel_on_interruption=False,
+    )
+
+    context = LLMContext([])
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
+        context,
+        user_params=LLMUserAggregatorParams(
+            user_turn_strategies=UserTurnStrategies(
+                stop=[SpeechTimeoutUserTurnStopStrategy()],
+            ),
+            vad_analyzer=SileroVADAnalyzer(),
+        ),
+    )
+
+    pipeline = Pipeline(
+        [
+            transport.input(),
+            user_aggregator,
+            llm,
+            transport.output(),
+            assistant_aggregator,
+        ]
+    )
+
+    task = PipelineTask(
+        pipeline,
+        params=PipelineParams(
+            enable_metrics=True,
+            enable_usage_metrics=True,
+        ),
+        idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
+    )
+
+    @transport.event_handler("on_client_connected")
+    async def on_client_connected(transport, client):
+        logger.info(f"Client connected")
+
+    @transport.event_handler("on_client_disconnected")
+    async def on_client_disconnected(transport, client):
+        logger.info(f"Client disconnected")
+        await task.cancel()
+
+    runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
+    await runner.run(task)
+
+
+async def bot(runner_args: RunnerArguments):
+    """Main bot entry point compatible with Pipecat Cloud."""
+    transport = await create_transport(runner_args, transport_params)
+    await run_bot(transport, runner_args)
+
+
+if __name__ == "__main__":
+    from pipecat.runner.run import main
+
+    main()

examples/transcription/transcription-gradium.py

@@ -51,7 +51,6 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
 
     stt = GradiumSTTService(
         api_key=os.environ["GRADIUM_API_KEY"],
-        api_endpoint_base_url="wss://us.api.gradium.ai/api/speech/asr",
         settings=GradiumSTTService.Settings(
             language=Language.EN,
             delay_in_frames=8,

examples/transcription/transcription-openai.py

@@ -49,13 +49,7 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = OpenAIRealtimeSTTService(
-        api_key=os.environ["OPENAI_API_KEY"],
-        settings=OpenAIRealtimeSTTService.Settings(
-            model="gpt-4o-transcribe",
-            prompt="Expect words related to dogs, such as breed names.",
-        ),
-    )
+    stt = OpenAIRealtimeSTTService(api_key=os.environ["OPENAI_API_KEY"])
 
     tl = TranscriptionLogger()
     vad_processor = VADProcessor(vad_analyzer=SileroVADAnalyzer())

examples/transports/transports-vonage.py

@@ -0,0 +1,134 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Example of using OpenAI Realtime voice LLM service with Vonage Video Connector transport."""
+
+import asyncio
+import os
+import sys
+from collections.abc import Callable
+from typing import Any
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.frames.frames import LLMRunFrame
+from pipecat.observers.loggers.transcription_log_observer import TranscriptionLogObserver
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.processors.aggregators.llm_response_universal import (
+    LLMContextAggregatorPair,
+    LLMUserAggregatorParams,
+)
+from pipecat.runner.vonage import configure
+from pipecat.services.openai.realtime.events import (
+    AudioConfiguration,
+    AudioInput,
+    InputAudioNoiseReduction,
+    InputAudioTranscription,
+    SemanticTurnDetection,
+    SessionProperties,
+)
+from pipecat.services.openai.realtime.llm import OpenAIRealtimeLLMService
+from pipecat.transports.vonage.video_connector import (
+    VonageVideoConnectorTransport,
+    VonageVideoConnectorTransportParams,
+)
+
+load_dotenv(override=True)
+
+logger.remove(0)
+logger.add(sys.stderr, level="DEBUG")
+
+
+async def main() -> None:
+    """Main entry point for the OpenAI Realtime vonage video connector example."""
+    (application_id, session_id, token) = await configure()
+
+    transport = VonageVideoConnectorTransport(
+        application_id,
+        session_id,
+        token,
+        VonageVideoConnectorTransportParams(
+            audio_in_enabled=True,
+            audio_out_enabled=True,
+            publisher_name="Bot",
+        ),
+    )
+
+    llm = OpenAIRealtimeLLMService(
+        api_key=os.environ["OPENAI_API_KEY"],
+        settings=OpenAIRealtimeLLMService.Settings(
+            system_instruction="""You are a helpful and friendly AI.
+
+Act like a human, but remember that you aren't a human and that you can't do human
+things in the real world. Your voice and personality should be warm and engaging, with a lively and
+playful tone.
+
+If interacting in a non-English language, start by using the standard accent or dialect familiar to
+the user. Talk quickly.
+
+You are participating in a voice conversation. Keep your responses concise, short, and to the point
+unless specifically asked to elaborate on a topic.
+
+Remember, your responses should be short. Just one or two sentences, usually. Respond in English.""",
+            session_properties=SessionProperties(
+                audio=AudioConfiguration(
+                    input=AudioInput(
+                        transcription=InputAudioTranscription(),
+                        turn_detection=SemanticTurnDetection(),
+                        noise_reduction=InputAudioNoiseReduction(type="near_field"),
+                    )
+                ),
+            ),
+        ),
+    )
+
+    context = LLMContext(
+        [{"role": "developer", "content": "Say hello!"}],
+    )
+
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
+        context,
+        user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()),
+    )
+
+    pipeline = Pipeline(
+        [
+            transport.input(),
+            user_aggregator,
+            llm,
+            transport.output(),
+            assistant_aggregator,
+        ]
+    )
+
+    task = PipelineTask(
+        pipeline,
+        params=PipelineParams(
+            enable_metrics=True,
+            enable_usage_metrics=True,
+        ),
+        observers=[TranscriptionLogObserver()],
+    )
+
+    event_handler: Callable[[str], Callable[[Any], Any]] = transport.event_handler
+
+    @event_handler("on_client_connected")
+    async def on_client_connected(transport: VonageVideoConnectorTransport, client: object) -> None:
+        logger.info("Client connected")
+        await task.queue_frames([LLMRunFrame()])
+
+    runner = PipelineRunner()
+
+    await runner.run(task)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

examples/turn-management/turn-management-filter-incomplete-turns-function-calling.py

@@ -0,0 +1,201 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Example 22: Filter Incomplete Turns
+
+Demonstrates LLM-based turn completion detection to suppress bot responses when
+the user was cut off mid-thought. The LLM outputs one of three markers:
+- ✓ (complete): User finished their thought, respond normally
+- ○ (incomplete short): User was cut off, wait ~5s then prompt
+- ◐ (incomplete long): User needs time to think, wait ~10s then prompt
+
+When incomplete is detected, the bot's response is suppressed. After the timeout
+expires, the LLM is automatically prompted to re-engage the user.
+"""
+
+import os
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.adapters.schemas.tools_schema import ToolsSchema
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.frames.frames import LLMRunFrame
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.processors.aggregators.llm_response_universal import (
+    AssistantTurnStoppedMessage,
+    LLMContextAggregatorPair,
+    LLMUserAggregatorParams,
+    UserTurnStoppedMessage,
+)
+from pipecat.runner.types import RunnerArguments
+from pipecat.runner.utils import create_transport
+from pipecat.services.cartesia.tts import CartesiaTTSService
+from pipecat.services.deepgram.stt import DeepgramSTTService
+from pipecat.services.llm_service import FunctionCallParams
+from pipecat.services.openai.llm import OpenAILLMService
+from pipecat.transports.base_transport import BaseTransport, TransportParams
+from pipecat.transports.daily.transport import DailyParams
+from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
+from pipecat.turns.user_turn_strategies import FilterIncompleteUserTurnStrategies
+
+load_dotenv(override=True)
+
+
+# We use lambdas to defer transport parameter creation until the transport
+# type is selected at runtime.
+transport_params = {
+    "daily": lambda: DailyParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "twilio": lambda: FastAPIWebsocketParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "webrtc": lambda: TransportParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+}
+
+
+async def get_weather(params: FunctionCallParams, location: str):
+    """Return the current weather for a location.
+
+    A stub that always reports the same conditions — replace with a real
+    weather API in production.
+
+    Args:
+        location (str): The city and state or country, e.g. "Paris, France".
+    """
+    await params.result_callback(
+        {
+            "location": location,
+            "temperature_celsius": 22,
+            "conditions": "partly cloudy",
+        }
+    )
+
+
+async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
+    logger.info(f"Starting bot")
+
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
+
+    llm = OpenAILLMService(
+        api_key=os.environ["OPENAI_API_KEY"],
+        settings=OpenAILLMService.Settings(
+            system_instruction=(
+                "You are a helpful assistant in a voice conversation. Your "
+                "responses will be spoken aloud, so avoid emojis, bullet "
+                "points, or other formatting that can't be spoken. Respond to "
+                "what the user said in a creative, helpful, and brief way. "
+                "If the user asks about the weather, call the get_weather "
+                "tool and speak the result back naturally."
+            ),
+        ),
+    )
+    llm.register_direct_function(get_weather)
+
+    tts = CartesiaTTSService(
+        api_key=os.environ["CARTESIA_API_KEY"],
+        settings=CartesiaTTSService.Settings(
+            voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
+        ),
+    )
+
+    context = LLMContext(tools=ToolsSchema(standard_tools=[get_weather]))
+    # `FilterIncompleteUserTurnStrategies` pairs the default detector
+    # chain with `LLMTurnCompletionUserTurnStopStrategy`: detectors
+    # trigger LLM inference but the public `on_user_turn_stopped` event
+    # fires only when the LLM confirms ✓. The LLM marks each response
+    # with one of:
+    # ✓ = complete (respond normally)
+    # ○ = incomplete short (wait 5s, then prompt)
+    # ◐ = incomplete long (wait 10s, then prompt)
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
+        context,
+        user_params=LLMUserAggregatorParams(
+            vad_analyzer=SileroVADAnalyzer(),
+            user_turn_strategies=FilterIncompleteUserTurnStrategies(
+                # Optional: customize turn completion behavior
+                # config=UserTurnCompletionConfig(
+                #     incomplete_short_timeout=5.0,
+                #     incomplete_long_timeout=10.0,
+                #     incomplete_short_prompt="Custom prompt...",
+                #     incomplete_long_prompt="Custom prompt...",
+                #     instructions="Custom turn completion instructions...",
+                # ),
+            ),
+        ),
+    )
+
+    pipeline = Pipeline(
+        [
+            transport.input(),  # Transport user input
+            stt,
+            user_aggregator,  # User responses
+            llm,  # LLM
+            tts,  # TTS
+            transport.output(),  # Transport bot output
+            assistant_aggregator,  # Assistant spoken responses
+        ]
+    )
+
+    task = PipelineTask(
+        pipeline,
+        params=PipelineParams(
+            enable_metrics=True,
+            enable_usage_metrics=True,
+        ),
+        idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
+    )
+
+    @transport.event_handler("on_client_connected")
+    async def on_client_connected(transport, client):
+        logger.info(f"Client connected")
+        # Kick off the conversation.
+        context.add_message(
+            {"role": "developer", "content": "Please introduce yourself to the user."}
+        )
+        await task.queue_frames([LLMRunFrame()])
+
+    @transport.event_handler("on_client_disconnected")
+    async def on_client_disconnected(transport, client):
+        logger.info(f"Client disconnected")
+        await task.cancel()
+
+    @user_aggregator.event_handler("on_user_turn_stopped")
+    async def on_user_turn_stopped(aggregator, strategy, message: UserTurnStoppedMessage):
+        timestamp = f"[{message.timestamp}] " if message.timestamp else ""
+        line = f"{timestamp}user: {message.content}"
+        logger.info(f"Transcript: {line}")
+
+    @assistant_aggregator.event_handler("on_assistant_turn_stopped")
+    async def on_assistant_turn_stopped(aggregator, message: AssistantTurnStoppedMessage):
+        timestamp = f"[{message.timestamp}] " if message.timestamp else ""
+        line = f"{timestamp}assistant: {message.content}"
+        logger.info(f"Transcript: {line}")
+
+    runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
+
+    await runner.run(task)
+
+
+async def bot(runner_args: RunnerArguments):
+    """Main bot entry point compatible with Pipecat Cloud."""
+    transport = await create_transport(runner_args, transport_params)
+    await run_bot(transport, runner_args)
+
+
+if __name__ == "__main__":
+    from pipecat.runner.run import main
+
+    main()

examples/turn-management/turn-management-filter-incomplete-turns.py

@@ -10,7 +10,7 @@ Demonstrates LLM-based turn completion detection to suppress bot responses when
 the user was cut off mid-thought. The LLM outputs one of three markers:
 - ✓ (complete): User finished their thought, respond normally
 - ○ (incomplete short): User was cut off, wait ~5s then prompt
-- ◐ (incomplete long): User needs time to think, wait ~15s then prompt
+- ◐ (incomplete long): User needs time to think, wait ~10s then prompt
 
 When incomplete is detected, the bot's response is suppressed. After the timeout
 expires, the LLM is automatically prompted to re-engage the user.
@@ -41,6 +41,7 @@ from pipecat.services.openai.llm import OpenAILLMService
 from pipecat.transports.base_transport import BaseTransport, TransportParams
 from pipecat.transports.daily.transport import DailyParams
 from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
+from pipecat.turns.user_turn_strategies import FilterIncompleteUserTurnStrategies
 
 load_dotenv(override=True)
 
@@ -83,23 +84,28 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     )
 
     context = LLMContext()
+    # `FilterIncompleteUserTurnStrategies` pairs the default detector
+    # chain with `LLMTurnCompletionUserTurnStopStrategy`: detectors
+    # trigger LLM inference but the public `on_user_turn_stopped` event
+    # fires only when the LLM confirms ✓. The LLM marks each response
+    # with one of:
+    # ✓ = complete (respond normally)
+    # ○ = incomplete short (wait 5s, then prompt)
+    # ◐ = incomplete long (wait 10s, then prompt)
     user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
         context,
         user_params=LLMUserAggregatorParams(
             vad_analyzer=SileroVADAnalyzer(),
-            # Enable turn completion filtering - the LLM will output:
-            # ✓ = complete (respond normally)
-            # ○ = incomplete short (wait 5s, then prompt)
-            # ◐ = incomplete long (wait 15s, then prompt)
-            filter_incomplete_user_turns=True,
-            # Optional: customize turn completion behavior
-            # turn_completion_config=TurnCompletionConfig(
-            #     incomplete_short_timeout=5.0,
-            #     incomplete_long_timeout=15.0,
-            #     incomplete_short_prompt="Custom prompt...",
-            #     incomplete_long_prompt="Custom prompt...",
-            #     instructions="Custom turn completion instructions...",
-            # ),
+            user_turn_strategies=FilterIncompleteUserTurnStrategies(
+                # Optional: customize turn completion behavior
+                # config=UserTurnCompletionConfig(
+                #     incomplete_short_timeout=5.0,
+                #     incomplete_long_timeout=10.0,
+                #     incomplete_short_prompt="Custom prompt...",
+                #     incomplete_long_prompt="Custom prompt...",
+                #     instructions="Custom turn completion instructions...",
+                # ),
+            ),
         ),
     )
 

examples/update-settings/stt/stt-gradium.py

@@ -50,10 +50,7 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = GradiumSTTService(
-        api_key=os.environ["GRADIUM_API_KEY"],
-        api_endpoint_base_url="wss://us.api.gradium.ai/api/speech/asr",
-    )
+    stt = GradiumSTTService(api_key=os.environ["GRADIUM_API_KEY"])
 
     tts = CartesiaTTSService(
         api_key=os.environ["CARTESIA_API_KEY"],

examples/update-settings/stt/stt-soniox.py

@@ -22,9 +22,9 @@ from pipecat.processors.aggregators.llm_response_universal import (
 )
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
-from pipecat.services.cartesia.tts import CartesiaTTSService
 from pipecat.services.openai.llm import OpenAILLMService
 from pipecat.services.soniox.stt import SonioxSTTService
+from pipecat.services.soniox.tts import SonioxTTSService
 from pipecat.transcriptions.language import Language
 from pipecat.transports.base_transport import BaseTransport, TransportParams
 from pipecat.transports.daily.transport import DailyParams
@@ -53,12 +53,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
 
     stt = SonioxSTTService(api_key=os.environ["SONIOX_API_KEY"])
 
-    tts = CartesiaTTSService(
-        api_key=os.environ["CARTESIA_API_KEY"],
-        settings=CartesiaTTSService.Settings(
-            voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
-        ),
-    )
+    tts = SonioxTTSService(api_key=os.environ["SONIOX_API_KEY"])
 
     llm = OpenAILLMService(
         api_key=os.environ["OPENAI_API_KEY"],
@@ -103,9 +98,9 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
         await task.queue_frames([LLMRunFrame()])
 
         await asyncio.sleep(10)
-        logger.info("Updating Soniox STT settings: language=es")
+        logger.info("Updating Soniox STT settings: language_hints=[es]")
         await task.queue_frame(
-            STTUpdateSettingsFrame(delta=SonioxSTTService.Settings(language=Language.ES))
+            STTUpdateSettingsFrame(delta=SonioxSTTService.Settings(language_hints=[Language.ES]))
         )
 
     @transport.event_handler("on_client_disconnected")

examples/update-settings/tts/tts-gradium.py

@@ -55,7 +55,6 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     tts = GradiumTTSService(
         api_key=os.environ["GRADIUM_API_KEY"],
         settings=GradiumTTSService.Settings(voice="YTpq7expH9539ERJ"),
-        url="wss://us.api.gradium.ai/api/speech/tts",
     )
 
     llm = OpenAILLMService(

examples/voice/voice-gradium.py

@@ -54,7 +54,6 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
 
     stt = GradiumSTTService(
         api_key=os.environ["GRADIUM_API_KEY"],
-        api_endpoint_base_url="wss://us.api.gradium.ai/api/speech/asr",
         settings=GradiumSTTService.Settings(
             language=Language.EN,
         ),
@@ -62,7 +61,6 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
 
     tts = GradiumTTSService(
         api_key=os.environ["GRADIUM_API_KEY"],
-        url="wss://us.api.gradium.ai/api/speech/tts",
         settings=GradiumTTSService.Settings(
             voice="YTpq7expH9539ERJ",
         ),

examples/voice/voice-nvidia-sagemaker.py

@@ -0,0 +1,129 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+# For a full example of how to deploy to SageMaker, see:
+# https://github.com/pipecat-ai/pipecat-examples/tree/main/nvidia_sagemaker_example/deployment/aws-sagemaker-nvidia
+
+import os
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.frames.frames import LLMRunFrame
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.processors.aggregators.llm_response_universal import (
+    LLMContextAggregatorPair,
+    LLMUserAggregatorParams,
+)
+from pipecat.runner.types import RunnerArguments
+from pipecat.runner.utils import create_transport
+from pipecat.services.nvidia.llm import NvidiaLLMService
+from pipecat.services.nvidia.sagemaker.stt import NvidiaSageMakerSTTService
+from pipecat.services.nvidia.sagemaker.tts import NvidiaSageMakerTTSService
+from pipecat.transports.base_transport import BaseTransport, TransportParams
+from pipecat.transports.daily.transport import DailyParams
+from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
+
+load_dotenv(override=True)
+
+# We use lambdas to defer transport parameter creation until the transport
+# type is selected at runtime.
+transport_params = {
+    "daily": lambda: DailyParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "twilio": lambda: FastAPIWebsocketParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+    "webrtc": lambda: TransportParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+    ),
+}
+
+
+async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
+    logger.info(f"Starting bot")
+
+    stt = NvidiaSageMakerSTTService(
+        endpoint_name=os.environ["SAGEMAKER_ASR_ENDPOINT_NAME"],
+        region=os.getenv("AWS_REGION", "us-west-2"),
+    )
+
+    llm = NvidiaLLMService(
+        api_key=os.environ["NVIDIA_API_KEY"],
+        settings=NvidiaLLMService.Settings(
+            model="meta/llama-3.3-70b-instruct",
+            system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
+        ),
+    )
+
+    tts = NvidiaSageMakerTTSService(
+        endpoint_name=os.environ["SAGEMAKER_MAGPIE_ENDPOINT_NAME"],
+        region=os.getenv("AWS_REGION", "us-west-2"),
+    )
+
+    context = LLMContext()
+    user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
+        context,
+        user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()),
+    )
+
+    pipeline = Pipeline(
+        [
+            transport.input(),  # Transport user input
+            stt,  # STT
+            user_aggregator,  # User responses
+            llm,  # LLM
+            tts,  # TTS
+            transport.output(),  # Transport bot output
+            assistant_aggregator,  # Assistant spoken responses
+        ]
+    )
+
+    task = PipelineTask(
+        pipeline,
+        params=PipelineParams(
+            enable_metrics=True,
+            enable_usage_metrics=True,
+        ),
+        idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
+    )
+
+    @transport.event_handler("on_client_connected")
+    async def on_client_connected(transport, client):
+        logger.info(f"Client connected")
+        # Kick off the conversation.
+        context.add_message(
+            {"role": "developer", "content": "Please introduce yourself to the user."}
+        )
+        await task.queue_frames([LLMRunFrame()])
+
+    @transport.event_handler("on_client_disconnected")
+    async def on_client_disconnected(transport, client):
+        logger.info(f"Client disconnected")
+        await task.cancel()
+
+    runner = PipelineRunner(handle_sigint=runner_args.handle_sigint)
+
+    await runner.run(task)
+
+
+async def bot(runner_args: RunnerArguments):
+    """Main bot entry point compatible with Pipecat Cloud."""
+    transport = await create_transport(runner_args, transport_params)
+    await run_bot(transport, runner_args)
+
+
+if __name__ == "__main__":
+    from pipecat.runner.run import main
+
+    main()

examples/voice/voice-openai.py

@@ -25,7 +25,6 @@ from pipecat.runner.utils import create_transport
 from pipecat.services.openai.llm import OpenAILLMService
 from pipecat.services.openai.stt import OpenAIRealtimeSTTService
 from pipecat.services.openai.tts import OpenAITTSService
-from pipecat.transcriptions.language import Language
 from pipecat.transports.base_transport import BaseTransport, TransportParams
 from pipecat.transports.daily.transport import DailyParams
 from pipecat.transports.websocket.fastapi import FastAPIWebsocketParams
@@ -53,14 +52,7 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = OpenAIRealtimeSTTService(
-        api_key=os.environ["OPENAI_API_KEY"],
-        settings=OpenAIRealtimeSTTService.Settings(
-            model="gpt-4o-transcribe",
-            prompt="Expect words related to dogs, such as breed names.",
-            language=Language.EN,
-        ),
-    )
+    stt = OpenAIRealtimeSTTService(api_key=os.environ["OPENAI_API_KEY"])
 
     tts = OpenAITTSService(
         api_key=os.environ["OPENAI_API_KEY"],
@@ -72,7 +64,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     llm = OpenAILLMService(
         api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAILLMService.Settings(
-            system_instruction="You are very knowledgable about dogs. Your output will be spoken aloud, so avoid special characters that can't easily be spoken, such as emojis or bullet points. Respond to what the user said in a creative and helpful way.",
+            system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),
     )
 

examples/voice/voice-soniox.py

@@ -58,6 +58,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
             # Add strict mode to enforce the language hints
             language_hints=[Language.EN],
             language_hints_strict=True,
+            enable_language_identification=True,
         ),
     )
 

pyproject.toml

@@ -77,6 +77,7 @@ groq = [ "groq>=0.23.0,<2" ]
 gstreamer = [ "pygobject~=3.50.0" ]
 heygen = [ "livekit>=1.0.13,<2", "pipecat-ai[websockets-base]" ]
 hume = [ "hume>=0.11.2,<1" ]
+inception = []
 inworld = [ "pipecat-ai[websockets-base]" ]
 koala = [ "pvkoala~=2.0.3" ]
 kokoro = [ "kokoro-onnx>=0.5.0,<1", "requests>=2.32.5,<3" ]
@@ -103,7 +104,7 @@ piper = [ "piper-tts>=1.3.0,<2", "requests>=2.32.5,<3" ]
 qwen = []
 resembleai = [ "pipecat-ai[websockets-base]" ]
 rime = [ "pipecat-ai[websockets-base]" ]
-runner = [ "python-dotenv>=1.0.0,<2.0.0", "uvicorn>=0.32.0,<1.0.0", "fastapi>=0.115.6,<1", "pipecat-ai-small-webrtc-prebuilt>=2.5.0"]
+runner = [ "python-dotenv>=1.0.0,<2.0.0", "uvicorn>=0.32.0,<1.0.0", "fastapi>=0.115.6,<1", "pipecat-ai-prebuilt>=1.0.1"]
 sagemaker = ["aws_sdk_sagemaker_runtime_http2; python_version>='3.12'"]
 sambanova = []
 sarvam = [ "sarvamai==0.1.28", "pipecat-ai[websockets-base]" ]
@@ -119,6 +120,7 @@ tavus = [ "pipecat-ai[daily]" ]
 together = []
 tracing = [ "opentelemetry-sdk>=1.33.0,<2", "opentelemetry-api>=1.33.0,<2", "opentelemetry-instrumentation>=0.54b0,<1" ]
 ultravox = [ "pipecat-ai[websockets-base]" ]
+vonage-video-connector = [ "vonage-video-connector~=0.2.3b0; python_full_version>='3.13' and python_full_version<'3.14' and platform_system=='Linux'" ]
 webrtc = [ "aiortc>=1.14.0,<2", "opencv-python>=4.11.0.86,<5" ]
 websocket = [ "pipecat-ai[websockets-base]", "fastapi>=0.115.6,<1" ]
 websockets-base = [ "websockets>=13.1,<16.0" ]

pyrightconfig.json

@@ -6,115 +6,54 @@
   "exclude": ["**/*_pb2.py", "**/__pycache__"],
   "ignore": [
     "tests",
-    "src/pipecat/adapters/services/anthropic_adapter.py",
-    "src/pipecat/adapters/services/aws_nova_sonic_adapter.py",
-    "src/pipecat/adapters/services/bedrock_adapter.py",
-    "src/pipecat/adapters/services/gemini_adapter.py",
-    "src/pipecat/adapters/services/grok_realtime_adapter.py",
-    "src/pipecat/adapters/services/inworld_realtime_adapter.py",
-    "src/pipecat/adapters/services/open_ai_adapter.py",
-    "src/pipecat/adapters/services/open_ai_realtime_adapter.py",
-    "src/pipecat/adapters/services/open_ai_responses_adapter.py",
-    "src/pipecat/adapters/services/perplexity_adapter.py",
-    "src/pipecat/audio/dtmf/utils.py",
     "src/pipecat/audio/filters/aic_filter.py",
     "src/pipecat/audio/filters/krisp_viva_filter.py",
-    "src/pipecat/audio/filters/rnnoise_filter.py",
     "src/pipecat/audio/turn/smart_turn/local_smart_turn_v2.py",
     "src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py",
     "src/pipecat/audio/vad/silero.py",
-    "src/pipecat/processors/aggregators/llm_context.py",
     "src/pipecat/processors/aggregators/llm_response_universal.py",
     "src/pipecat/processors/frame_processor.py",
-    "src/pipecat/processors/frameworks/langchain.py",
     "src/pipecat/processors/frameworks/rtvi/observer.py",
-    "src/pipecat/processors/frameworks/rtvi/processor.py",
-    "src/pipecat/processors/frameworks/strands_agents.py",
     "src/pipecat/services/anthropic/llm.py",
-    "src/pipecat/services/assemblyai/stt.py",
-    "src/pipecat/services/aws/agent_core.py",
     "src/pipecat/services/aws/llm.py",
     "src/pipecat/services/aws/nova_sonic/llm.py",
     "src/pipecat/services/aws/sagemaker/bidi_client.py",
-    "src/pipecat/services/aws/stt.py",
-    "src/pipecat/services/aws/tts.py",
-    "src/pipecat/services/aws/utils.py",
-    "src/pipecat/services/azure/stt.py",
     "src/pipecat/services/azure/tts.py",
-    "src/pipecat/services/cartesia/stt.py",
-    "src/pipecat/services/deepgram/flux/base.py",
     "src/pipecat/services/deepgram/flux/sagemaker/stt.py",
-    "src/pipecat/services/deepgram/flux/stt.py",
     "src/pipecat/services/deepgram/sagemaker/stt.py",
     "src/pipecat/services/deepgram/sagemaker/tts.py",
-    "src/pipecat/services/deepgram/tts.py",
-    "src/pipecat/services/elevenlabs/stt.py",
-    "src/pipecat/services/elevenlabs/tts.py",
-    "src/pipecat/services/fish/tts.py",
-    "src/pipecat/services/gladia/stt.py",
     "src/pipecat/services/google/gemini_live/llm.py",
-    "src/pipecat/services/google/gemini_live/vertex/llm.py",
-    "src/pipecat/services/google/image.py",
     "src/pipecat/services/google/llm.py",
     "src/pipecat/services/google/stt.py",
     "src/pipecat/services/google/tts.py",
-    "src/pipecat/services/gradium/stt.py",
-    "src/pipecat/services/groq/tts.py",
-    "src/pipecat/services/heygen/api_interactive_avatar.py",
-    "src/pipecat/services/heygen/base_api.py",
     "src/pipecat/services/heygen/client.py",
     "src/pipecat/services/heygen/video.py",
-    "src/pipecat/services/hume/tts.py",
     "src/pipecat/services/inworld/realtime/llm.py",
-    "src/pipecat/services/inworld/tts.py",
-    "src/pipecat/services/kokoro/tts.py",
     "src/pipecat/services/llm_service.py",
-    "src/pipecat/services/lmnt/tts.py",
     "src/pipecat/services/mem0/memory.py",
-    "src/pipecat/services/mistral/stt.py",
     "src/pipecat/services/mistral/tts.py",
-    "src/pipecat/services/moondream/vision.py",
-    "src/pipecat/services/neuphonic/tts.py",
     "src/pipecat/services/nvidia/stt.py",
     "src/pipecat/services/nvidia/tts.py",
     "src/pipecat/services/openai/base_llm.py",
-    "src/pipecat/services/openai/image.py",
-    "src/pipecat/services/openai/llm.py",
     "src/pipecat/services/openai/realtime/llm.py",
-    "src/pipecat/services/openai/responses/llm.py",
-    "src/pipecat/services/openai/stt.py",
-    "src/pipecat/services/openai/tts.py",
-    "src/pipecat/services/openrouter/llm.py",
-    "src/pipecat/services/piper/tts.py",
-    "src/pipecat/services/resembleai/tts.py",
     "src/pipecat/services/rime/tts.py",
     "src/pipecat/services/sambanova/llm.py",
     "src/pipecat/services/sarvam/stt.py",
-    "src/pipecat/services/sarvam/tts.py",
     "src/pipecat/services/simli/video.py",
-    "src/pipecat/services/smallest/tts.py",
-    "src/pipecat/services/soniox/stt.py",
     "src/pipecat/services/speechmatics/stt.py",
     "src/pipecat/services/stt_service.py",
     "src/pipecat/services/tavus/video.py",
     "src/pipecat/services/tts_service.py",
     "src/pipecat/services/ultravox/llm.py",
-    "src/pipecat/services/websocket_service.py",
     "src/pipecat/services/whisper/stt.py",
     "src/pipecat/services/xai/realtime/llm.py",
-    "src/pipecat/services/xtts/tts.py",
-    "src/pipecat/transports/base_output.py",
     "src/pipecat/transports/daily/transport.py",
-    "src/pipecat/transports/heygen/transport.py",
     "src/pipecat/transports/lemonslice/transport.py",
     "src/pipecat/transports/livekit/transport.py",
     "src/pipecat/transports/smallwebrtc/connection.py",
-    "src/pipecat/transports/smallwebrtc/request_handler.py",
     "src/pipecat/transports/smallwebrtc/transport.py",
     "src/pipecat/transports/tavus/transport.py",
-    "src/pipecat/transports/websocket/client.py",
-    "src/pipecat/transports/websocket/server.py",
-    "src/pipecat/transports/whatsapp/client.py"
+    "src/pipecat/transports/websocket/server.py"
   ],
   "reportMissingImports": false
 }

scripts/evals/run-release-evals.py

@@ -198,6 +198,7 @@ TESTS_FUNCTION_CALLING = [
     ("function-calling/function-calling-sarvam.py", EVAL_WEATHER),
     ("function-calling/function-calling-novita.py", EVAL_WEATHER),
     ("function-calling/function-calling-deepseek.py", EVAL_WEATHER),
+    ("function-calling/function-calling-inception.py", EVAL_WEATHER),
     # Video
     ("function-calling/function-calling-anthropic-video.py", EVAL_VISION_CAMERA),
     ("function-calling/function-calling-aws-video.py", EVAL_VISION_CAMERA),
@@ -223,12 +224,11 @@ TESTS_REALTIME = [
     # ("realtime/realtime-azure.py", EVAL_WEATHER),
     ("realtime/realtime-openai-text.py", EVAL_WEATHER),
     ("realtime/realtime-openai-live-video.py", EVAL_VISION_CAMERA),
-    ("realtime/realtime-gemini-live.py", EVAL_SIMPLE_MATH),
+    ("realtime/realtime-gemini-live.py", EVAL_WEATHER),
     ("realtime/realtime-gemini-live-local-vad.py", EVAL_SIMPLE_MATH),
-    ("realtime/realtime-gemini-live-function-calling.py", EVAL_WEATHER),
     ("realtime/realtime-gemini-live-video.py", EVAL_VISION_CAMERA),
     ("realtime/realtime-gemini-live-google-search.py", EVAL_ONLINE_SEARCH),
-    ("realtime/realtime-gemini-live-vertex-function-calling.py", EVAL_WEATHER),
+    ("realtime/realtime-gemini-live-vertex.py", EVAL_WEATHER),
     ("realtime/realtime-aws-nova-sonic.py", EVAL_SIMPLE_MATH),
     ("realtime/realtime-ultravox.py", EVAL_ORDER),
     ("realtime/realtime-grok.py", EVAL_WEATHER),
@@ -243,6 +243,7 @@ TESTS_VIDEO_AVATAR = [
 
 TESTS_TURN_MANAGEMENT = [
     ("turn-management/turn-management-filter-incomplete-turns.py", EVAL_COMPLETE_TURN),
+    ("turn-management/turn-management-filter-incomplete-turns-function-calling.py", EVAL_WEATHER),
 ]
 
 TESTS_THINKING = [

src/pipecat/adapters/services/anthropic_adapter.py

@@ -9,7 +9,7 @@
 import copy
 import json
 from dataclasses import dataclass
-from typing import Any, TypedDict, TypeGuard, TypeVar
+from typing import Any, TypedDict, TypeGuard, TypeVar, cast
 
 from anthropic import NOT_GIVEN, NotGiven
 from anthropic.types.message_param import MessageParam
@@ -121,16 +121,20 @@ class AnthropicLLMAdapter(BaseLLMAdapter[AnthropicLLMInvocationParams]):
         messages = self._from_universal_context_messages(self.get_messages(context)).messages
 
         # Sanitize messages for logging
-        messages_for_logging = []
+        messages_for_logging: list[dict[str, Any]] = []
         for message in messages:
-            msg = copy.deepcopy(message)
-            if "content" in msg:
-                if isinstance(msg["content"], list):
-                    for item in msg["content"]:
-                        if item["type"] == "image":
-                            item["source"]["data"] = "..."
-                        if item["type"] == "thinking" and item.get("signature"):
-                            item["signature"] = "..."
+            msg: dict[str, Any] = copy.deepcopy(dict(message))
+            content = msg.get("content")
+            if isinstance(content, list):
+                for item in content:
+                    if not isinstance(item, dict):
+                        continue
+                    if item.get("type") == "image":
+                        source = item.get("source")
+                        if isinstance(source, dict):
+                            source["data"] = "..."
+                    if item.get("type") == "thinking" and item.get("signature"):
+                        item["signature"] = "..."
             messages_for_logging.append(msg)
         return messages_for_logging
 
@@ -185,8 +189,13 @@ class AnthropicLLMAdapter(BaseLLMAdapter[AnthropicLLMInvocationParams]):
                     ]
                 if isinstance(next_message["content"], str):
                     next_message["content"] = [{"type": "text", "text": next_message["content"]}]
-                # Concatenate the content
-                current_message["content"].extend(next_message["content"])
+                # Concatenate the content. MessageParam types content as
+                # `str | Iterable[...]`, but this codebase assumes it's
+                # either a str or a list. The str case is handled above, so
+                # we assume that both are lists here.
+                cast(list[Any], current_message["content"]).extend(
+                    cast(list[Any], next_message["content"])
+                )
                 # Remove the next message from the list
                 messages.pop(i + 1)
             else:
@@ -239,7 +248,7 @@ class AnthropicLLMAdapter(BaseLLMAdapter[AnthropicLLMInvocationParams]):
             }
 
         # Fall back to assuming that the message is already in Anthropic format
-        return copy.deepcopy(message.message)
+        return cast(MessageParam, copy.deepcopy(message.message))
 
     def _from_standard_message(self, message: LLMStandardMessage) -> MessageParam:
         """Convert standard universal context message to Anthropic format.
@@ -280,20 +289,26 @@ class AnthropicLLMAdapter(BaseLLMAdapter[AnthropicLLMInvocationParams]):
                     ]
                 }
         """
-        message = copy.deepcopy(message)
-        if message["role"] == "tool":
-            return {
-                "role": "user",
-                "content": [
-                    {
-                        "type": "tool_result",
-                        "tool_use_id": message["tool_call_id"],
-                        "content": message["content"],
-                    },
-                ],
-            }
-        if message.get("tool_calls"):
-            tc = message["tool_calls"]
+        # ChatCompletionMessageParam (input) and MessageParam (output) are
+        # different TypedDicts — work with the message as a plain dict for the
+        # transformations below and cast back to MessageParam at return sites.
+        msg = cast(dict[str, Any], copy.deepcopy(message))
+        if msg["role"] == "tool":
+            return cast(
+                MessageParam,
+                {
+                    "role": "user",
+                    "content": [
+                        {
+                            "type": "tool_result",
+                            "tool_use_id": msg["tool_call_id"],
+                            "content": msg["content"],
+                        },
+                    ],
+                },
+            )
+        if msg.get("tool_calls"):
+            tc = msg["tool_calls"]
             ret = {"role": "assistant", "content": []}
             for tool_call in tc:
                 function = tool_call["function"]
@@ -305,8 +320,8 @@ class AnthropicLLMAdapter(BaseLLMAdapter[AnthropicLLMInvocationParams]):
                     "input": arguments,
                 }
                 ret["content"].append(new_tool_use)
-            return ret
-        content = message.get("content")
+            return cast(MessageParam, ret)
+        content = msg.get("content")
         if isinstance(content, str):
             # fix empty text
             if content == "":
@@ -354,7 +369,7 @@ class AnthropicLLMAdapter(BaseLLMAdapter[AnthropicLLMInvocationParams]):
                     image_item = content.pop(img_idx)
                     content.insert(first_txt_idx, image_item)
 
-        return message
+        return cast(MessageParam, msg)
 
     def _with_cache_control_markers(self, messages: list[MessageParam]) -> list[MessageParam]:
         """Add cache control markers to messages for prompt caching.
@@ -369,7 +384,16 @@ class AnthropicLLMAdapter(BaseLLMAdapter[AnthropicLLMInvocationParams]):
         def add_cache_control_marker(message: MessageParam):
             if isinstance(message["content"], str):
                 message["content"] = [{"type": "text", "text": message["content"]}]
-            message["content"][-1]["cache_control"] = {"type": "ephemeral"}
+            # Assumptions on the next line:
+            #   - content is a list (str case handled above; this codebase only
+            #     ever constructs content as a str or a list)
+            #   - the list is non-empty (guaranteed by the empty-content
+            #     replacement in `_from_universal_context_messages`)
+            #   - the last item is a dict. The standard-message path enforces
+            #     this via TypedDicts (which are dicts at runtime); the
+            #     LLMSpecificMessage passthrough doesn't, but in practice
+            #     callers use dicts.
+            cast(list[Any], message["content"])[-1]["cache_control"] = {"type": "ephemeral"}
 
         try:
             # Add cache control markers to the most recent two user messages.

src/pipecat/adapters/services/aws_nova_sonic_adapter.py

@@ -8,9 +8,9 @@
 
 import copy
 import json
-from dataclasses import dataclass
+from dataclasses import asdict, dataclass
 from enum import Enum
-from typing import Any, TypedDict
+from typing import Any, TypedDict, cast
 
 from loguru import logger
 
@@ -110,7 +110,10 @@ class AWSNovaSonicLLMAdapter(BaseLLMAdapter[AWSNovaSonicLLMInvocationParams]):
         Returns:
             List of messages in a format ready for logging about AWS Nova Sonic.
         """
-        return self._from_universal_context_messages(self.get_messages(context)).messages
+        return [
+            asdict(m)
+            for m in self._from_universal_context_messages(self.get_messages(context)).messages
+        ]
 
     @dataclass
     class ConvertedMessages:
@@ -123,18 +126,27 @@ class AWSNovaSonicLLMAdapter(BaseLLMAdapter[AWSNovaSonicLLMInvocationParams]):
         self, universal_context_messages: list[LLMContextMessage]
     ) -> ConvertedMessages:
         system_instruction = None
-        messages = []
+        messages: list[AWSNovaSonicConversationHistoryMessage] = []
 
         # Bail if there are no messages
         if not universal_context_messages:
-            return self.ConvertedMessages()
+            return self.ConvertedMessages(messages=[])
 
-        universal_context_messages = copy.deepcopy(universal_context_messages)
+        # NOTE: This adapter does not yet handle ``LLMSpecificMessage`` —
+        # those are filtered out below (the role-extraction and conversion
+        # logic only applies to standard message dicts). If/when this
+        # adapter grows a per-provider passthrough like the Anthropic
+        # adapter has, LLMSpecific items can flow through.
+        ucm: list[dict[str, Any]] = [
+            cast(dict[str, Any], m)
+            for m in copy.deepcopy(universal_context_messages)
+            if isinstance(m, dict)
+        ]
 
         # If we have a "system" message as our first message,
         # pull that out into "instruction"
-        if universal_context_messages[0].get("role") == "system":
-            system = universal_context_messages.pop(0)
+        if ucm and ucm[0].get("role") == "system":
+            system = ucm.pop(0)
             content = system.get("content")
             if isinstance(content, str):
                 system_instruction = content
@@ -145,19 +157,21 @@ class AWSNovaSonicLLMAdapter(BaseLLMAdapter[AWSNovaSonicLLMInvocationParams]):
 
         # Convert any remaining "system"/"developer" messages to "user",
         # as Nova Sonic only supports "user" and "assistant" in history.
-        for msg in universal_context_messages:
+        for msg in ucm:
             if msg.get("role") in ("system", "developer"):
                 msg["role"] = "user"
 
         # Process remaining messages to fill out conversation history.
-        for universal_context_message in universal_context_messages:
+        for universal_context_message in ucm:
             message = self._from_universal_context_message(universal_context_message)
             if message:
                 messages.append(message)
 
         return self.ConvertedMessages(messages=messages, system_instruction=system_instruction)
 
-    def _from_universal_context_message(self, message) -> AWSNovaSonicConversationHistoryMessage:
+    def _from_universal_context_message(
+        self, message: dict[str, Any]
+    ) -> AWSNovaSonicConversationHistoryMessage | None:
         """Convert standard message format to Nova Sonic format.
 
         Args:
@@ -167,17 +181,18 @@ class AWSNovaSonicLLMAdapter(BaseLLMAdapter[AWSNovaSonicLLMInvocationParams]):
             Nova Sonic conversation history message, or None if not convertible.
         """
         role = message.get("role")
-        if message.get("role") == "user" or message.get("role") == "assistant":
+        if role == "user" or role == "assistant":
             content = message.get("content")
-            if isinstance(message.get("content"), list):
-                content = ""
-                for c in message.get("content"):
+            if isinstance(content, list):
+                text_parts = []
+                for c in content:
                     if c.get("type") == "text":
-                        content += " " + c.get("text")
+                        text_parts.append(c.get("text"))
                     else:
                         logger.error(
                             f"Unhandled content type in context message: {c.get('type')} - {message}"
                         )
+                content = " ".join(t for t in text_parts if t)
             # There won't be content if this is an assistant tool call entry.
             # We're ignoring those since they can't be loaded into AWS Nova Sonic conversation
             # history

src/pipecat/adapters/services/bedrock_adapter.py

@@ -10,7 +10,7 @@ import base64
 import copy
 import json
 from dataclasses import dataclass
-from typing import Any, TypedDict
+from typing import Any, TypedDict, cast
 
 from loguru import logger
 
@@ -68,16 +68,19 @@ class AWSBedrockLLMAdapter(BaseLLMAdapter[AWSBedrockLLMInvocationParams]):
             system_instruction,
             discard_context_system=True,
         )
-        return {
-            "system": [{"text": effective_system}] if effective_system else None,
-            "messages": converted.messages,
-            # NOTE: LLMContext's tools are guaranteed to be a ToolsSchema (or NOT_GIVEN)
-            "tools": self.from_standard_tools(context.tools) or [],
-            # To avoid refactoring in AWSBedrockLLMService, we just pass through tool_choice.
-            # Eventually (when we don't have to maintain the non-LLMContext code path) we should do
-            # the conversion to Bedrock's expected format here rather than in AWSBedrockLLMService.
-            "tool_choice": context.tool_choice,
-        }
+        return cast(
+            AWSBedrockLLMInvocationParams,
+            {
+                "system": [{"text": effective_system}] if effective_system else None,
+                "messages": converted.messages,
+                # NOTE: LLMContext's tools are guaranteed to be a ToolsSchema (or NOT_GIVEN)
+                "tools": self.from_standard_tools(context.tools) or [],
+                # To avoid refactoring in AWSBedrockLLMService, we just pass through tool_choice.
+                # Eventually (when we don't have to maintain the non-LLMContext code path) we should do
+                # the conversion to Bedrock's expected format here rather than in AWSBedrockLLMService.
+                "tool_choice": context.tool_choice,
+            },
+        )
 
     def get_messages_for_logging(self, context) -> list[dict[str, Any]]:
         """Get messages from a universal LLM context in a format ready for logging about AWS Bedrock.
@@ -213,35 +216,36 @@ class AWSBedrockLLMAdapter(BaseLLMAdapter[AWSBedrockLLMInvocationParams]):
                     ]
                 }
         """
-        message = copy.deepcopy(message)
-        if message["role"] == "tool":
+        # ChatCompletionMessageParam (input) and the dict shape Bedrock expects
+        # are different — work with the deepcopied message as a plain dict for
+        # the transformations below.
+        msg = cast(dict[str, Any], copy.deepcopy(message))
+        if msg["role"] == "tool":
             # Try to parse the content as JSON if it looks like JSON
             try:
-                if message["content"].strip().startswith("{") and message[
-                    "content"
-                ].strip().endswith("}"):
-                    content_json = json.loads(message["content"])
+                if msg["content"].strip().startswith("{") and msg["content"].strip().endswith("}"):
+                    content_json = json.loads(msg["content"])
                     tool_result_content = [{"json": content_json}]
                 else:
-                    tool_result_content = [{"text": message["content"]}]
+                    tool_result_content = [{"text": msg["content"]}]
             except (json.JSONDecodeError, ValueError, AttributeError):
-                tool_result_content = [{"text": message["content"]}]
+                tool_result_content = [{"text": msg["content"]}]
 
             return {
                 "role": "user",
                 "content": [
                     {
                         "toolResult": {
-                            "toolUseId": message["tool_call_id"],
+                            "toolUseId": msg["tool_call_id"],
                             "content": tool_result_content,
                         },
                     },
                 ],
             }
 
-        if message.get("tool_calls"):
-            tc = message["tool_calls"]
-            ret = {"role": "assistant", "content": []}
+        if msg.get("tool_calls"):
+            tc = msg["tool_calls"]
+            ret: dict[str, Any] = {"role": "assistant", "content": []}
             for tool_call in tc:
                 function = tool_call["function"]
                 arguments = json.loads(function["arguments"])
@@ -256,12 +260,12 @@ class AWSBedrockLLMAdapter(BaseLLMAdapter[AWSBedrockLLMInvocationParams]):
             return ret
 
         # Handle text content
-        content = message.get("content")
+        content = msg.get("content")
         if isinstance(content, str):
             if content == "":
-                return {"role": message["role"], "content": [{"text": "(empty)"}]}
+                return {"role": msg["role"], "content": [{"text": "(empty)"}]}
             else:
-                return {"role": message["role"], "content": [{"text": content}]}
+                return {"role": msg["role"], "content": [{"text": content}]}
         elif isinstance(content, list):
             new_content = []
             for item in content:
@@ -300,9 +304,9 @@ class AWSBedrockLLMAdapter(BaseLLMAdapter[AWSBedrockLLMInvocationParams]):
                     # Move image before the first text
                     image_item = new_content.pop(img_idx)
                 new_content.insert(first_txt_idx, image_item)
-            return {"role": message["role"], "content": new_content}
+            return {"role": msg["role"], "content": new_content}
 
-        return message
+        return msg
 
     @staticmethod
     def _to_bedrock_function_format(function: FunctionSchema) -> dict[str, Any]:

src/pipecat/adapters/services/gemini_adapter.py

@@ -9,7 +9,7 @@
 import base64
 import json
 from dataclasses import dataclass, field
-from typing import Any, TypedDict
+from typing import Any, TypedDict, cast
 
 from loguru import logger
 from openai import NotGiven
@@ -139,6 +139,36 @@ class GeminiLLMAdapter(BaseLLMAdapter[GeminiLLMInvocationParams]):
 
         return formatted_standard_tools + custom_gemini_tools
 
+    @staticmethod
+    def to_function_response_dict(content: Any) -> dict[str, Any]:
+        """Convert a tool-result content value to Gemini's FunctionResponse.response shape.
+
+        Gemini's ``FunctionResponse.response`` field requires a dict, so
+        non-dict values (e.g. plain strings, JSON-encoded scalars, or
+        sentinel strings like ``"COMPLETED"`` used when a function returned
+        no value) are wrapped as ``{"value": <value>}``. JSON strings that
+        decode to a dict are passed through as-is.
+
+        Args:
+            content: The tool-result content. Typically the JSON-encoded
+                return value of a function, but can also be a plain string
+                (e.g. ``"COMPLETED"``) or already-parsed dict.
+
+        Returns:
+            A dict suitable for ``FunctionResponse.response``.
+        """
+        if isinstance(content, dict):
+            return content
+        if not isinstance(content, str):
+            return {"value": content}
+        try:
+            decoded = json.loads(content)
+        except (json.JSONDecodeError, ValueError):
+            return {"value": content}
+        if isinstance(decoded, dict):
+            return decoded
+        return {"value": decoded}
+
     def get_messages_for_logging(self, context: LLMContext) -> list[dict[str, Any]]:
         """Get messages from a universal LLM context in a format ready for logging about Gemini.
 
@@ -154,9 +184,12 @@ class GeminiLLMAdapter(BaseLLMAdapter[GeminiLLMInvocationParams]):
         messages = self._from_universal_context_messages(self.get_messages(context)).messages
 
         # Sanitize messages for logging
-        messages_for_logging = []
+        messages_for_logging: list[dict[str, Any]] = []
         for message in messages:
-            obj = message.to_json_dict()
+            # `to_json_dict()` returns `dict[str, object]`; treat as a plain
+            # dict for the value indexing/mutation below. The broad `except`
+            # below is the safety net if any item isn't shaped as expected.
+            obj: dict[str, Any] = cast(dict[str, Any], message.to_json_dict())
             try:
                 if "parts" in obj:
                     for part in obj["parts"]:
@@ -274,7 +307,8 @@ class GeminiLLMAdapter(BaseLLMAdapter[GeminiLLMInvocationParams]):
         # Check if we only have function-related messages (no regular text)
         effective_system = extracted_system or system_instruction
         has_regular_messages = any(
-            len(msg.parts) == 1
+            msg.parts is not None
+            and len(msg.parts) == 1
             and getattr(msg.parts[0], "text", None)
             and not getattr(msg.parts[0], "function_call", None)
             and not getattr(msg.parts[0], "function_response", None)
@@ -346,8 +380,11 @@ class GeminiLLMAdapter(BaseLLMAdapter[GeminiLLMInvocationParams]):
                     parts=[Part(function_call=FunctionCall(name="search", args={"query": "test"}))]
                 )
         """
-        role = message["role"]
-        content = message.get("content", [])
+        # ChatCompletionMessageParam (a union of TypedDicts) doesn't allow
+        # the dict-style key access used below; treat it as a plain dict.
+        msg = cast(dict[str, Any], message)
+        role = msg["role"]
+        content = msg.get("content", [])
 
         # Convert non-initial system/developer messages to user role,
         # as Gemini doesn't support these as input messages.
@@ -359,8 +396,8 @@ class GeminiLLMAdapter(BaseLLMAdapter[GeminiLLMInvocationParams]):
         parts = []
         tool_call_id_to_name_mapping = {}
 
-        if message.get("tool_calls"):
-            for tc in message["tool_calls"]:
+        if msg.get("tool_calls"):
+            for tc in msg["tool_calls"]:
                 id = tc["id"]
                 name = tc["function"]["name"]
                 tool_call_id_to_name_mapping[id] = name
@@ -375,19 +412,10 @@ class GeminiLLMAdapter(BaseLLMAdapter[GeminiLLMInvocationParams]):
                 )
         elif role == "tool":
             role = "user"
-            try:
-                response = json.loads(message["content"])
-                if isinstance(response, dict):
-                    response_dict = response
-                else:
-                    response_dict = {"value": response}
-            except Exception as e:
-                # Response might not be JSON-deserializable.
-                # This occurs with a UserImageFrame, for example, where we get a plain "COMPLETED" string.
-                response_dict = {"value": message["content"]}
+            response_dict = self.to_function_response_dict(msg["content"])
 
             # Get function name from mapping using tool_call_id, or fallback
-            tool_call_id = message.get("tool_call_id")
+            tool_call_id = msg.get("tool_call_id")
             function_name = "tool_call_result"  # Default fallback
             if tool_call_id and tool_call_id in params.tool_call_id_to_name_mapping:
                 function_name = params.tool_call_id_to_name_mapping[tool_call_id]
@@ -491,7 +519,7 @@ class GeminiLLMAdapter(BaseLLMAdapter[GeminiLLMInvocationParams]):
 
         def is_tool_call_message(msg: Content) -> bool:
             """Check if message contains only function_call parts."""
-            return (
+            return bool(
                 msg.role == "model"
                 and msg.parts
                 and all(getattr(part, "function_call", None) for part in msg.parts)
@@ -499,6 +527,8 @@ class GeminiLLMAdapter(BaseLLMAdapter[GeminiLLMInvocationParams]):
 
         def message_has_thought_signature(msg: Content) -> bool:
             """Check if any part in the message has a thought_signature."""
+            if msg.parts is None:
+                return False
             return any(getattr(part, "thought_signature", None) for part in msg.parts)
 
         merged_messages = []
@@ -564,6 +594,8 @@ class GeminiLLMAdapter(BaseLLMAdapter[GeminiLLMInvocationParams]):
         logger.debug(f"Thought signatures to apply: {len(thought_signature_dicts)}")
         for ts in thought_signature_dicts:
             bookmark = ts.get("bookmark")
+            if bookmark is None:
+                continue
             if bookmark.get("function_call"):
                 logger.trace(f" - To function call: {bookmark['function_call']}")
             elif bookmark.get("text"):
@@ -665,6 +697,8 @@ class GeminiLLMAdapter(BaseLLMAdapter[GeminiLLMInvocationParams]):
         if (
             hasattr(part, "inline_data")
             and part.inline_data
+            and part.inline_data.data is not None
+            and bookmark_inline_data.data is not None
             # Comparing length should be good enough for matching inline data,
             # especially since we're already matching thought signatures in
             # strict message order. Comparing actual data is expensive.

src/pipecat/adapters/services/grok_realtime_adapter.py

@@ -13,7 +13,7 @@ Grok's Voice Agent API.
 import copy
 import json
 from dataclasses import dataclass
-from typing import Any, TypedDict
+from typing import Any, TypedDict, cast
 
 from loguru import logger
 
@@ -85,7 +85,10 @@ class GrokRealtimeLLMAdapter(BaseLLMAdapter):
         Returns:
             List of messages with sensitive data redacted.
         """
-        return self.get_messages(context, truncate_large_values=True)
+        return cast(
+            list[dict[str, Any]],
+            self.get_messages(context, truncate_large_values=True),
+        )
 
     @dataclass
     class ConvertedMessages:
@@ -111,11 +114,20 @@ class GrokRealtimeLLMAdapter(BaseLLMAdapter):
         if not universal_context_messages:
             return self.ConvertedMessages(messages=[])
 
-        messages = copy.deepcopy(universal_context_messages)
+        # NOTE: This adapter does not yet handle ``LLMSpecificMessage`` —
+        # those are filtered out below. Other adapters (e.g. Anthropic)
+        # dispatch LLMSpecific items through a per-provider passthrough.
+        # The pack-into-single-text-message strategy here doesn't compose
+        # with opaque per-provider payloads.
+        messages: list[dict[str, Any]] = [
+            cast(dict[str, Any], m)
+            for m in copy.deepcopy(universal_context_messages)
+            if isinstance(m, dict)
+        ]
         system_instruction = None
 
         # Extract system message as session instructions
-        if messages[0].get("role") == "system":
+        if messages and messages[0].get("role") == "system":
             system = messages.pop(0)
             content = system.get("content")
             if isinstance(content, str):
@@ -133,7 +145,9 @@ class GrokRealtimeLLMAdapter(BaseLLMAdapter):
         # Single user message can be sent normally
         if len(messages) == 1 and messages[0].get("role") == "user":
             return self.ConvertedMessages(
-                messages=[self._from_universal_context_message(messages[0])],
+                messages=[
+                    self._from_universal_context_message(cast(LLMContextMessage, messages[0]))
+                ],
                 system_instruction=system_instruction,
             )
 
@@ -181,26 +195,29 @@ class GrokRealtimeLLMAdapter(BaseLLMAdapter):
         Returns:
             ConversationItem formatted for Grok Realtime API.
         """
-        if message.get("role") == "user":
-            content = message.get("content")
+        # NOTE: ``LLMSpecificMessage`` is not yet handled here — see the
+        # corresponding note in `_from_universal_context_messages`.
+        msg = cast(dict[str, Any], message)
+        if msg.get("role") == "user":
+            content = msg.get("content")
             if isinstance(content, list):
-                text_content = ""
+                text_parts = []
                 for c in content:
                     if c.get("type") == "text":
-                        text_content += " " + c.get("text")
+                        text_parts.append(c.get("text"))
                     else:
                         logger.error(
-                            f"Unhandled content type in context message: {c.get('type')} - {message}"
+                            f"Unhandled content type in context message: {c.get('type')} - {msg}"
                         )
-                content = text_content.strip()
+                content = " ".join(t for t in text_parts if t).strip()
             return events.ConversationItem(
                 role="user",
                 type="message",
                 content=[events.ItemContent(type="input_text", text=content)],
             )
 
-        if message.get("role") == "assistant" and message.get("tool_calls"):
-            tc = message.get("tool_calls")[0]
+        if msg.get("role") == "assistant" and msg.get("tool_calls"):
+            tc = msg["tool_calls"][0]
             return events.ConversationItem(
                 type="function_call",
                 call_id=tc["id"],
@@ -208,7 +225,7 @@ class GrokRealtimeLLMAdapter(BaseLLMAdapter):
                 arguments=tc["function"]["arguments"],
             )
 
-        logger.error(f"Unhandled message type in _from_universal_context_message: {message}")
+        raise ValueError(f"Unhandled message type in _from_universal_context_message: {msg}")
 
     @staticmethod
     def _to_grok_function_format(function: FunctionSchema) -> dict[str, Any]:

src/pipecat/adapters/services/inworld_realtime_adapter.py

@@ -13,7 +13,7 @@ Inworld's Realtime API.
 import copy
 import json
 from dataclasses import dataclass
-from typing import Any, TypedDict
+from typing import Any, TypedDict, cast
 
 from loguru import logger
 
@@ -85,7 +85,10 @@ class InworldRealtimeLLMAdapter(BaseLLMAdapter):
         Returns:
             List of messages with sensitive data redacted.
         """
-        return self.get_messages(context, truncate_large_values=True)
+        return cast(
+            list[dict[str, Any]],
+            self.get_messages(context, truncate_large_values=True),
+        )
 
     @dataclass
     class ConvertedMessages:
@@ -111,11 +114,20 @@ class InworldRealtimeLLMAdapter(BaseLLMAdapter):
         if not universal_context_messages:
             return self.ConvertedMessages(messages=[])
 
-        messages = copy.deepcopy(universal_context_messages)
+        # NOTE: This adapter does not yet handle ``LLMSpecificMessage`` —
+        # those are filtered out below. Other adapters (e.g. Anthropic)
+        # dispatch LLMSpecific items through a per-provider passthrough.
+        # The pack-into-single-text-message strategy here doesn't compose
+        # with opaque per-provider payloads.
+        messages: list[dict[str, Any]] = [
+            cast(dict[str, Any], m)
+            for m in copy.deepcopy(universal_context_messages)
+            if isinstance(m, dict)
+        ]
         system_instruction = None
 
         # Extract system message as session instructions
-        if messages[0].get("role") == "system":
+        if messages and messages[0].get("role") == "system":
             system = messages.pop(0)
             content = system.get("content")
             if isinstance(content, str):
@@ -133,7 +145,9 @@ class InworldRealtimeLLMAdapter(BaseLLMAdapter):
         # Single user message can be sent normally
         if len(messages) == 1 and messages[0].get("role") == "user":
             return self.ConvertedMessages(
-                messages=[self._from_universal_context_message(messages[0])],
+                messages=[
+                    self._from_universal_context_message(cast(LLMContextMessage, messages[0]))
+                ],
                 system_instruction=system_instruction,
             )
 
@@ -181,26 +195,29 @@ class InworldRealtimeLLMAdapter(BaseLLMAdapter):
         Returns:
             ConversationItem formatted for Inworld Realtime API.
         """
-        if message.get("role") == "user":
-            content = message.get("content")
+        # NOTE: ``LLMSpecificMessage`` is not yet handled here — see the
+        # corresponding note in `_from_universal_context_messages`.
+        msg = cast(dict[str, Any], message)
+        if msg.get("role") == "user":
+            content = msg.get("content")
             if isinstance(content, list):
-                text_content = ""
+                text_parts = []
                 for c in content:
                     if c.get("type") == "text":
-                        text_content += " " + c.get("text")
+                        text_parts.append(c.get("text"))
                     else:
                         logger.error(
-                            f"Unhandled content type in context message: {c.get('type')} - {message}"
+                            f"Unhandled content type in context message: {c.get('type')} - {msg}"
                         )
-                content = text_content.strip()
+                content = " ".join(t for t in text_parts if t).strip()
             return events.ConversationItem(
                 role="user",
                 type="message",
                 content=[events.ItemContent(type="input_text", text=content)],
             )
 
-        if message.get("role") == "assistant" and message.get("tool_calls"):
-            tc = message.get("tool_calls")[0]
+        if msg.get("role") == "assistant" and msg.get("tool_calls"):
+            tc = msg["tool_calls"][0]
             return events.ConversationItem(
                 type="function_call",
                 call_id=tc["id"],
@@ -208,7 +225,7 @@ class InworldRealtimeLLMAdapter(BaseLLMAdapter):
                 arguments=tc["function"]["arguments"],
             )
 
-        logger.error(f"Unhandled message type in _from_universal_context_message: {message}")
+        raise ValueError(f"Unhandled message type in _from_universal_context_message: {msg}")
 
     @staticmethod
     def _to_inworld_function_format(function: FunctionSchema) -> dict[str, Any]:

src/pipecat/adapters/services/open_ai_adapter.py

@@ -127,12 +127,15 @@ class OpenAILLMAdapter(BaseLLMAdapter[OpenAILLMInvocationParams]):
         )
 
         if system_instruction:
-            # Detect initial system message for warning purposes (don't extract)
-            initial_content = (
-                messages[0].get("content", "")
-                if messages and messages[0].get("role") == "system"
-                else None
-            )
+            # Detect initial system message for warning purposes (don't extract).
+            # ChatCompletionMessageParam.content is `str | Iterable[...]`; we
+            # only forward it for warning purposes, so coerce non-strings to
+            # None — the resolver handles None.
+            initial_content: str | None = None
+            if messages and messages[0].get("role") == "system":
+                raw_content = messages[0].get("content", "")
+                if isinstance(raw_content, str):
+                    initial_content = raw_content
             self._resolve_system_instruction(
                 initial_content,
                 system_instruction,
@@ -140,12 +143,15 @@ class OpenAILLMAdapter(BaseLLMAdapter[OpenAILLMInvocationParams]):
             )
             messages = [{"role": "system", "content": system_instruction}] + messages
 
-        return {
-            "messages": messages,
-            # NOTE; LLMContext's tools are guaranteed to be a ToolsSchema (or NOT_GIVEN)
-            "tools": self.from_standard_tools(context.tools),
-            "tool_choice": _openai_from_llm_context_tool_choice(context.tool_choice),
-        }
+        return cast(
+            OpenAILLMInvocationParams,
+            {
+                "messages": messages,
+                # NOTE; LLMContext's tools are guaranteed to be a ToolsSchema (or NOT_GIVEN)
+                "tools": self.from_standard_tools(context.tools),
+                "tool_choice": _openai_from_llm_context_tool_choice(context.tool_choice),
+            },
+        )
 
     def to_provider_tools_format(self, tools_schema: ToolsSchema) -> list[ChatCompletionToolParam]:
         """Convert function schemas to OpenAI's function-calling format.
@@ -158,13 +164,19 @@ class OpenAILLMAdapter(BaseLLMAdapter[OpenAILLMInvocationParams]):
             with ChatCompletion API.
         """
         functions_schema = tools_schema.standard_tools
-        formatted_standard_tools = [
-            ChatCompletionToolParam(type="function", function=func.to_default_dict())
+        # `function=...` expects a `FunctionDefinition` TypedDict; the dict
+        # produced by `to_default_dict()` is structurally compatible. Cast at
+        # the boundary.
+        formatted_standard_tools: list[ChatCompletionToolParam] = [
+            ChatCompletionToolParam(type="function", function=cast(Any, func.to_default_dict()))
             for func in functions_schema
         ]
-        custom_openai_tools = []
+        custom_openai_tools: list[ChatCompletionToolParam] = []
         if tools_schema.custom_tools:
-            custom_openai_tools = tools_schema.custom_tools.get(AdapterType.OPENAI, [])
+            custom_openai_tools = cast(
+                list[ChatCompletionToolParam],
+                tools_schema.custom_tools.get(AdapterType.OPENAI, []),
+            )
         return formatted_standard_tools + custom_openai_tools
 
     def get_messages_for_logging(self, context: LLMContext) -> list[dict[str, Any]]:
@@ -178,7 +190,10 @@ class OpenAILLMAdapter(BaseLLMAdapter[OpenAILLMInvocationParams]):
         Returns:
             List of messages in a format ready for logging about OpenAI.
         """
-        return self.get_messages(context, truncate_large_values=True)
+        return cast(
+            list[dict[str, Any]],
+            self.get_messages(context, truncate_large_values=True),
+        )
 
     def _from_universal_context_messages(
         self,

src/pipecat/adapters/services/open_ai_realtime_adapter.py

@@ -9,7 +9,7 @@
 import copy
 import json
 from dataclasses import dataclass
-from typing import Any, TypedDict
+from typing import Any, TypedDict, cast
 
 from loguru import logger
 
@@ -81,7 +81,7 @@ class OpenAIRealtimeLLMAdapter(BaseLLMAdapter):
         Returns:
             List of messages in a format ready for logging about OpenAI Realtime.
         """
-        return self.get_messages(context, truncate_large_values=True)
+        return cast(list[dict[str, Any]], self.get_messages(context, truncate_large_values=True))
 
     @dataclass
     class ConvertedMessages:
@@ -101,12 +101,24 @@ class OpenAIRealtimeLLMAdapter(BaseLLMAdapter):
         if not universal_context_messages:
             return self.ConvertedMessages(messages=[])
 
-        messages = copy.deepcopy(universal_context_messages)
+        # NOTE: This adapter does not yet handle ``LLMSpecificMessage`` — those
+        # are filtered out below. Other adapters (e.g. Anthropic) dispatch
+        # LLMSpecific items through a per-provider passthrough. For OpenAI
+        # Realtime, the strategy here packs a multi-message history into a
+        # single text message (see comment further down), which doesn't
+        # compose with opaque per-provider payloads. If/when this adapter
+        # adopts the per-message strategy, LLMSpecific items can flow
+        # through `_from_universal_context_message` like in other adapters.
+        messages: list[dict[str, Any]] = [
+            cast(dict[str, Any], m)
+            for m in copy.deepcopy(universal_context_messages)
+            if isinstance(m, dict)
+        ]
         system_instruction = None
 
         # If we have a "system" message as our first message,
         # pull that out into session "instructions"
-        if messages[0].get("role") == "system":
+        if messages and messages[0].get("role") == "system":
             system = messages.pop(0)
             content = system.get("content")
             if isinstance(content, str):
@@ -124,7 +136,9 @@ class OpenAIRealtimeLLMAdapter(BaseLLMAdapter):
         # If we have just a single "user" item, we can just send it normally
         if len(messages) == 1 and messages[0].get("role") == "user":
             return self.ConvertedMessages(
-                messages=[self._from_universal_context_message(messages[0])],
+                messages=[
+                    self._from_universal_context_message(cast(LLMContextMessage, messages[0]))
+                ],
                 system_instruction=system_instruction,
             )
 
@@ -142,18 +156,18 @@ class OpenAIRealtimeLLMAdapter(BaseLLMAdapter):
 
         return self.ConvertedMessages(
             messages=[
-                {
-                    "role": "user",
-                    "type": "message",
-                    "content": [
-                        {
-                            "type": "input_text",
-                            "text": "\n\n".join(
+                events.ConversationItem(
+                    role="user",
+                    type="message",
+                    content=[
+                        events.ItemContent(
+                            type="input_text",
+                            text="\n\n".join(
                                 [intro_text, json.dumps(messages, indent=2), trailing_text]
                             ),
-                        }
+                        )
                     ],
-                }
+                )
             ],
             system_instruction=system_instruction,
         )
@@ -161,31 +175,34 @@ class OpenAIRealtimeLLMAdapter(BaseLLMAdapter):
     def _from_universal_context_message(
         self, message: LLMContextMessage
     ) -> events.ConversationItem:
-        if message.get("role") == "user":
-            content = message.get("content")
-            if isinstance(message.get("content"), list):
+        # NOTE: ``LLMSpecificMessage`` is not yet handled here — see the
+        # corresponding note in `_from_universal_context_messages`.
+        msg = cast(dict[str, Any], message)
+        if msg.get("role") == "user":
+            content = msg.get("content")
+            if isinstance(content, list):
                 content = ""
-                for c in message.get("content"):
+                for c in msg.get("content", []):
                     if c.get("type") == "text":
                         content += " " + c.get("text")
                     else:
                         logger.error(
-                            f"Unhandled content type in context message: {c.get('type')} - {message}"
+                            f"Unhandled content type in context message: {c.get('type')} - {msg}"
                         )
             return events.ConversationItem(
                 role="user",
                 type="message",
                 content=[events.ItemContent(type="input_text", text=content)],
             )
-        if message.get("role") == "assistant" and message.get("tool_calls"):
-            tc = message.get("tool_calls")[0]
+        if msg.get("role") == "assistant" and msg.get("tool_calls"):
+            tc = msg["tool_calls"][0]
             return events.ConversationItem(
                 type="function_call",
                 call_id=tc["id"],
                 name=tc["function"]["name"],
                 arguments=tc["function"]["arguments"],
             )
-        logger.error(f"Unhandled message type in _from_universal_context_message: {message}")
+        raise ValueError(f"Unhandled message type in _from_universal_context_message: {msg}")
 
     @staticmethod
     def _to_openai_realtime_function_format(function: FunctionSchema) -> dict[str, Any]:

src/pipecat/adapters/services/open_ai_responses_adapter.py

@@ -6,7 +6,7 @@
 
 """OpenAI Responses API adapter for Pipecat."""
 
-from typing import Any, TypedDict
+from typing import Any, Required, TypedDict, cast
 
 from openai._types import NotGiven as OpenAINotGiven
 from openai.types.responses import FunctionToolParam, ResponseInputItemParam, ToolParam
@@ -23,8 +23,10 @@ from pipecat.processors.aggregators.llm_context import (
 class OpenAIResponsesLLMInvocationParams(TypedDict, total=False):
     """Context-based parameters for invoking OpenAI Responses API."""
 
-    input: list[ResponseInputItemParam]
-    tools: list[ToolParam] | OpenAINotGiven
+    # `input` and `tools` are always populated by `get_llm_invocation_params`;
+    # `instructions` is only set when a system instruction is present.
+    input: Required[list[ResponseInputItemParam]]
+    tools: Required[list[ToolParam] | OpenAINotGiven]
     instructions: str
 
 
@@ -64,8 +66,11 @@ class OpenAIResponsesLLMAdapter(BaseLLMAdapter[OpenAIResponsesLLMInvocationParam
         if system_instruction and messages:
             first_msg = messages[0] if not isinstance(messages[0], LLMSpecificMessage) else None
             if first_msg and first_msg.get("role") == "system":
+                # `content` is `str | Iterable[...]`; we only forward it for
+                # warning purposes. Coerce non-strings to None.
+                first_content = first_msg.get("content", "")
                 self._resolve_system_instruction(
-                    first_msg.get("content", ""),
+                    first_content if isinstance(first_content, str) else None,
                     system_instruction,
                     discard_context_system=False,
                 )
@@ -143,7 +148,10 @@ class OpenAIResponsesLLMAdapter(BaseLLMAdapter[OpenAIResponsesLLMInvocationParam
         Returns:
             List of messages in a format ready for logging.
         """
-        return self.get_messages(context, truncate_large_values=True)
+        return cast(
+            list[dict[str, Any]],
+            self.get_messages(context, truncate_large_values=True),
+        )
 
     def _convert_messages_to_input(
         self, messages: list[LLMContextMessage]
@@ -169,13 +177,15 @@ class OpenAIResponsesLLMAdapter(BaseLLMAdapter[OpenAIResponsesLLMInvocationParam
                 content = message.get("content", "")
                 if isinstance(content, list):
                     content = self._convert_multimodal_content(content)
-                result.append({"role": "developer", "content": content})
+                result.append(
+                    cast(ResponseInputItemParam, {"role": "developer", "content": content})
+                )
 
             elif role == "user":
                 content = message.get("content", "")
                 if isinstance(content, list):
                     content = self._convert_multimodal_content(content)
-                result.append({"role": "user", "content": content})
+                result.append(cast(ResponseInputItemParam, {"role": "user", "content": content}))
 
             elif role == "assistant":
                 tool_calls = message.get("tool_calls")
@@ -194,7 +204,9 @@ class OpenAIResponsesLLMAdapter(BaseLLMAdapter[OpenAIResponsesLLMInvocationParam
                     content = message.get("content", "")
                     if isinstance(content, list):
                         content = self._convert_multimodal_content(content)
-                    result.append({"role": "assistant", "content": content})
+                    result.append(
+                        cast(ResponseInputItemParam, {"role": "assistant", "content": content})
+                    )
 
             elif role == "tool":
                 content = message.get("content", "")

src/pipecat/adapters/services/perplexity_adapter.py

@@ -28,6 +28,7 @@ the messages are sent to Perplexity's API.
 """
 
 import copy
+from typing import Any, cast
 
 from openai.types.chat import ChatCompletionMessageParam
 
@@ -116,7 +117,11 @@ class PerplexityLLMAdapter(OpenAILLMAdapter):
         if not messages:
             return messages
 
-        messages = copy.deepcopy(messages)
+        # ChatCompletionMessageParam is a union of TypedDicts; the
+        # transformations below mutate by key/index in ways those TypedDicts
+        # don't permit. Work against a plain-dict view for the duration of
+        # the transformation and cast back at the return site.
+        msgs: list[dict[str, Any]] = cast(list[dict[str, Any]], copy.deepcopy(messages))
 
         # Note: "developer" → "user" conversion is handled by the parent adapter
         # via the convert_developer_to_user parameter.
@@ -125,10 +130,10 @@ class PerplexityLLMAdapter(OpenAILLMAdapter):
         # Perplexity allows system messages at the start, but rejects them
         # after any non-system message.
         in_initial_system_block = True
-        for i in range(len(messages)):
-            if messages[i].get("role") == "system":
+        for i in range(len(msgs)):
+            if msgs[i].get("role") == "system":
                 if not in_initial_system_block:
-                    messages[i]["role"] = "user"
+                    msgs[i]["role"] = "user"
             else:
                 in_initial_system_block = False
 
@@ -137,9 +142,9 @@ class PerplexityLLMAdapter(OpenAILLMAdapter):
         # messages that violate Perplexity's strict alternation requirement.
         # Skip consecutive system messages at the start — Perplexity allows those.
         i = 0
-        while i < len(messages) - 1:
-            current = messages[i]
-            next_msg = messages[i + 1]
+        while i < len(msgs) - 1:
+            current = msgs[i]
+            next_msg = msgs[i + 1]
             if current["role"] == next_msg["role"] == "system":
                 # Perplexity allows multiple initial system messages, don't merge
                 i += 1
@@ -154,7 +159,7 @@ class PerplexityLLMAdapter(OpenAILLMAdapter):
                     next_msg.get("content"), list
                 ):
                     current["content"].extend(next_msg["content"])
-                messages.pop(i + 1)
+                msgs.pop(i + 1)
             else:
                 i += 1
 
@@ -162,7 +167,7 @@ class PerplexityLLMAdapter(OpenAILLMAdapter):
         # Perplexity requires the last message to be "user" or "tool".
         # OpenAI appears to silently ignore trailing assistant messages
         # server-side, so removing them preserves equivalent behavior.
-        while messages and messages[-1].get("role") == "assistant":
-            messages.pop()
+        while msgs and msgs[-1].get("role") == "assistant":
+            msgs.pop()
 
-        return messages
+        return cast(list[ChatCompletionMessageParam], msgs)

src/pipecat/audio/dtmf/utils.py

@@ -14,7 +14,7 @@ in-memory after first load to improve performance on subsequent accesses.
 import asyncio
 import io
 import wave
-from importlib.resources import files
+from importlib.resources import as_file, files
 
 import aiofiles
 
@@ -52,10 +52,12 @@ async def load_dtmf_audio(button: KeypadEntry, *, sample_rate: int = 8000) -> by
             __DTMF_RESAMPLER__ = create_file_resampler()
 
         dtmf_file_name = __DTMF_FILE_NAME.get(button, f"dtmf-{button.value}.wav")
-        dtmf_file_path = files("pipecat.audio.dtmf").joinpath(dtmf_file_name)
-
-        async with aiofiles.open(dtmf_file_path, "rb") as f:
-            data = await f.read()
+        # `as_file` materialises the resource as a real filesystem `Path`,
+        # which aiofiles can open. (For installed packages this is just the
+        # bundled file; for zipped distributions it would extract to a temp.)
+        with as_file(files("pipecat.audio.dtmf").joinpath(dtmf_file_name)) as dtmf_file_path:
+            async with aiofiles.open(dtmf_file_path, "rb") as f:
+                data = await f.read()
 
         with io.BytesIO(data) as buffer:
             with wave.open(buffer, "rb") as wf:

src/pipecat/audio/filters/rnnoise_filter.py

@@ -60,7 +60,12 @@ class RNNoiseFilter(BaseAudioFilter):
         self._sample_rate = sample_rate
 
         try:
-            # RNNoise always requires 48kHz
+            # The module-level import sets `RNNoise` to `None` if pyrnnoise
+            # isn't installed; raise instead of calling `None(...)` so the
+            # except clause handles it cleanly.
+            if RNNoise is None:
+                raise ImportError("pyrnnoise is not installed")
+            # RNNoise always requires 48kHz.
             self._rnnoise = RNNoise(sample_rate=48000)
             self._rnnoise_ready = True
         except Exception as e:
@@ -107,7 +112,7 @@ class RNNoiseFilter(BaseAudioFilter):
         Returns:
             Noise-suppressed audio data as bytes.
         """
-        if not self._rnnoise_ready or not self._filtering:
+        if not self._rnnoise_ready or not self._filtering or self._rnnoise is None:
             return audio
 
         # Resample input if needed

src/pipecat/audio/resamplers/resampy_resampler.py

@@ -10,6 +10,8 @@ This module provides an audio resampler that uses the resampy library
 for high-quality audio sample rate conversion.
 """
 
+import warnings
+
 import numpy as np
 import resampy
 
@@ -21,6 +23,11 @@ class ResampyResampler(BaseAudioResampler):
 
     This resampler uses the resampy library's Kaiser windowing filter
     for high-quality audio resampling with good performance characteristics.
+
+    .. deprecated:: 1.2.0
+        ResampyResampler is deprecated and will be removed in Pipecat 2.0.
+        Use SOXRAudioResampler, create_file_resampler(), or create_stream_resampler()
+        instead.
     """
 
     def __init__(self, **kwargs):
@@ -29,7 +36,15 @@ class ResampyResampler(BaseAudioResampler):
         Args:
             **kwargs: Additional keyword arguments (currently unused).
         """
-        pass
+        with warnings.catch_warnings():
+            warnings.simplefilter("always")
+            warnings.warn(
+                "ResampyResampler is deprecated and will be removed in Pipecat 2.0. "
+                "Use SOXRAudioResampler, create_file_resampler(), or "
+                "create_stream_resampler() instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
 
     async def resample(self, audio: bytes, in_rate: int, out_rate: int) -> bytes:
         """Resample audio data using resampy library.

src/pipecat/frames/frames.py

@@ -339,6 +339,40 @@ class LLMTextFrame(TextFrame):
         self.includes_inter_frame_spaces = True
 
 
+@dataclass
+class LLMMarkerFrame(DataFrame):
+    """Sideband marker emitted by an LLM service.
+
+    A marker is short, structured assistant output that should be
+    persisted in the conversation context but should not flow through
+    the standard text path (TTS, transcript). The assistant aggregator
+    writes the marker to the context so the LLM can self-condition on
+    prior markers on subsequent turns.
+
+    The primary use today is the ``filter_incomplete_user_turns``
+    protocol, where ``UserTurnCompletionLLMServiceMixin`` emits the
+    turn-completion markers ✓ / ○ / ◐ on every response. The frame is
+    intentionally generic so other components — STT services with
+    built-in turn signals, end-of-turn classifiers, custom annotations,
+    etc. — can use the same mechanism to inject sideband signals into
+    the assistant context.
+
+    Parameters:
+        marker: The marker payload (typically a short string such as a
+            single character).
+        append_to_context_immediately: If True, the marker is written
+            to the context as its own standalone assistant message as
+            soon as it's received. If False, the marker is appended to
+            the running assistant aggregation and flushed to the
+            context together with the following text as a single
+            message (e.g. for the ✓ case the context message ends up
+            as "✓ <response>").
+    """
+
+    marker: str
+    append_to_context_immediately: bool = True
+
+
 @dataclass
 class AggregatedTextFrame(TextFrame):
     """Text frame representing an aggregation of TextFrames.
@@ -349,10 +383,14 @@ class AggregatedTextFrame(TextFrame):
     Parameters:
         aggregated_by: Method used to aggregate the text frames.
         context_id: Unique identifier for the TTS context that generated this text.
+        raw_text: The full matched text including start/end pattern delimiters, set when
+            this frame was produced from a PatternMatch (e.g. a ``<code>...</code>`` block).
+            None for ordinary sentence aggregations.
     """
 
     aggregated_by: AggregationType | str
     context_id: str | None = None
+    raw_text: str | None = None
 
 
 @dataclass
@@ -661,6 +699,11 @@ class FunctionCallResultProperties:
         is_final: Whether this is the final result for the function call. When
             ``False`` the result is treated as an intermediate update. Defaults to ``True``.
             Only meaningful for async function calls (``cancel_on_interruption=False``).
+            Note: realtime LLM services do not support streamed intermediate
+            results; they deliver only the final result to the provider. An
+            intermediate result reported to a realtime service is dropped
+            and an error is raised. Use a non-realtime LLM service if your
+            tool needs to stream intermediate results.
     """
 
     run_llm: bool | None = None
@@ -970,6 +1013,24 @@ class UserSpeakingFrame(SystemFrame):
     pass
 
 
+@dataclass
+class UserTurnInferenceCompletedFrame(SystemFrame):
+    """Frame indicating that the user turn is semantically complete.
+
+    Emitted by any component that can judge conversational turn
+    completeness — for example an LLM with turn-completion markers, an
+    STT service with built-in turn detection, or a dedicated
+    end-of-turn classifier. Stop strategies that gate the
+    user-turn-stop event on an external completeness signal (e.g.
+    ``LLMTurnCompletionUserTurnStopStrategy``) consume this frame to
+    finalize the turn. Producers should emit this frame only when they
+    judge the turn complete; an absence of this frame means the turn is
+    not yet considered complete.
+    """
+
+    pass
+
+
 @dataclass
 class VADUserStartedSpeakingFrame(SystemFrame):
     """Frame emitted when VAD definitively detects user started speaking.

src/pipecat/pipeline/task.py

@@ -303,7 +303,7 @@ class PipelineTask(BasePipelineTask):
 
         # This task maneger will handle all the asyncio tasks created by this
         # PipelineTask and its frame processors.
-        self._task_manager = task_manager or TaskManager()
+        self._pipeline_task_manager = task_manager or TaskManager()
 
         # This queue is the queue used to push frames to the pipeline.
         self._push_queue = asyncio.Queue()
@@ -386,7 +386,7 @@ class PipelineTask(BasePipelineTask):
         # The task observer acts as a proxy to the provided observers. This way,
         # we only need to pass a single observer (using the StartFrame) which
         # then just acts as a proxy.
-        self._observer = TaskObserver(observers=observers, task_manager=self._task_manager)
+        self._observer = TaskObserver(observers=observers)
 
         # These events can be used to check which frames make it to the source
         # or sink processors. Instead of calling the event handlers for every
@@ -654,32 +654,24 @@ class PipelineTask(BasePipelineTask):
 
     async def _create_tasks(self):
         """Create and start all pipeline processing tasks."""
-        self._process_push_task = self._task_manager.create_task(
-            self._process_push_queue(), f"{self}::_process_push_queue"
-        )
+        self._process_push_task = self.create_task(self._process_push_queue())
         return self._process_push_task
 
     def _maybe_start_heartbeat_tasks(self):
         """Start heartbeat tasks if heartbeats are enabled and not already running."""
         if self._params.enable_heartbeats and self._heartbeat_push_task is None:
-            self._heartbeat_push_task = self._task_manager.create_task(
-                self._heartbeat_push_handler(), f"{self}::_heartbeat_push_handler"
-            )
-            self._heartbeat_monitor_task = self._task_manager.create_task(
-                self._heartbeat_monitor_handler(), f"{self}::_heartbeat_monitor_handler"
-            )
+            self._heartbeat_push_task = self.create_task(self._heartbeat_push_handler())
+            self._heartbeat_monitor_task = self.create_task(self._heartbeat_monitor_handler())
 
     def _maybe_start_idle_task(self):
         """Start idle monitoring task if idle timeout is configured."""
         if self._idle_timeout_secs:
-            self._idle_monitor_task = self._task_manager.create_task(
-                self._idle_monitor_handler(), f"{self}::_idle_monitor_handler"
-            )
+            self._idle_monitor_task = self.create_task(self._idle_monitor_handler())
 
     async def _cancel_tasks(self):
         """Cancel all running pipeline tasks."""
         if self._process_push_task:
-            await self._task_manager.cancel_task(self._process_push_task)
+            await self.cancel_task(self._process_push_task)
             self._process_push_task = None
 
         await self._maybe_cancel_heartbeat_tasks()
@@ -691,17 +683,17 @@ class PipelineTask(BasePipelineTask):
             return
 
         if self._heartbeat_push_task:
-            await self._task_manager.cancel_task(self._heartbeat_push_task)
+            await self.cancel_task(self._heartbeat_push_task)
             self._heartbeat_push_task = None
 
         if self._heartbeat_monitor_task:
-            await self._task_manager.cancel_task(self._heartbeat_monitor_task)
+            await self.cancel_task(self._heartbeat_monitor_task)
             self._heartbeat_monitor_task = None
 
     async def _maybe_cancel_idle_task(self):
         """Cancel idle monitoring task if it is running."""
         if self._idle_monitor_task:
-            await self._task_manager.cancel_task(self._idle_monitor_task)
+            await self.cancel_task(self._idle_monitor_task)
             self._idle_monitor_task = None
 
     def _initial_metrics_frame(self) -> MetricsFrame:
@@ -759,12 +751,14 @@ class PipelineTask(BasePipelineTask):
 
     async def _setup(self, params: PipelineTaskParams):
         """Set up the pipeline task and all processors."""
+        await super().setup(self._pipeline_task_manager)
+
         mgr_params = TaskManagerParams(loop=params.loop)
-        self._task_manager.setup(mgr_params)
+        self.task_manager.setup(mgr_params)
 
         setup = FrameProcessorSetup(
             clock=self._clock,
-            task_manager=self._task_manager,
+            task_manager=self.task_manager,
             observer=self._observer,
             pipeline_task=self,
             # Populate the deprecated `tool_resources` field for backwards
@@ -780,6 +774,7 @@ class PipelineTask(BasePipelineTask):
         await self._load_setup_files()
 
         # Start task observer.
+        await self._observer.setup(self.task_manager)
         await self._observer.start()
 
     async def _cleanup(self, cleanup_pipeline: bool):
@@ -1019,7 +1014,7 @@ class PipelineTask(BasePipelineTask):
 
     def _print_dangling_tasks(self):
         """Log any dangling tasks that haven't been properly cleaned up."""
-        tasks = [t.get_name() for t in self._task_manager.current_tasks()]
+        tasks = [t.get_name() for t in self.task_manager.current_tasks()]
         if tasks:
             logger.warning(f"{self} dangling tasks detected: {tasks}")
 

src/pipecat/pipeline/task_observer.py

@@ -17,7 +17,6 @@ from typing import Any
 from attr import dataclass
 
 from pipecat.observers.base_observer import BaseObserver, FrameProcessed, FramePushed
-from pipecat.utils.asyncio.task_manager import BaseTaskManager
 
 
 @dataclass
@@ -62,19 +61,16 @@ class TaskObserver(BaseObserver):
         self,
         *,
         observers: list[BaseObserver] | None = None,
-        task_manager: BaseTaskManager,
         **kwargs,
     ):
         """Initialize the TaskObserver.
 
         Args:
             observers: List of observers to manage. Defaults to empty list.
-            task_manager: Task manager for creating and managing observer tasks.
             **kwargs: Additional arguments passed to the base observer.
         """
         super().__init__(**kwargs)
         self._observers = observers or []
-        self._task_manager = task_manager
         self._proxies: dict[BaseObserver, Proxy] | None = (
             None  # Becomes a dict after start() is called
         )
@@ -106,7 +102,7 @@ class TaskObserver(BaseObserver):
             # Remove the proxy so it doesn't get called anymore.
             del self._proxies[observer]
             # Cancel the proxy task right away.
-            await self._task_manager.cancel_task(proxy.task)
+            await self.cancel_task(proxy.task)
 
         # Remove the observer from the list.
         if observer in self._observers:
@@ -122,7 +118,7 @@ class TaskObserver(BaseObserver):
             return
 
         for proxy in self._proxies.values():
-            await self._task_manager.cancel_task(proxy.task)
+            await self.cancel_task(proxy.task)
 
     async def cleanup(self):
         """Cleanup all proxy observers."""
@@ -157,9 +153,8 @@ class TaskObserver(BaseObserver):
     def _create_proxy(self, observer: BaseObserver) -> Proxy:
         """Create a proxy for a single observer."""
         queue = asyncio.Queue()
-        task = self._task_manager.create_task(
-            self._proxy_task_handler(queue, observer),
-            f"TaskObserver::{observer}::_proxy_task_handler",
+        task = self.create_task(
+            self._proxy_task_handler(queue, observer), f"{observer}::_proxy_task_handler"
         )
         proxy = Proxy(queue=queue, task=task, observer=observer)
         return proxy

src/pipecat/processors/aggregators/async_tool_messages.py

@@ -0,0 +1,286 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Helpers for the async-tool message protocol used in LLM contexts.
+
+When a function is registered with ``cancel_on_interruption=False``, the
+``LLMUserContextAggregator`` / ``LLMAssistantContextAggregator`` pair appends
+async-tool messages to the conversation context as the underlying task
+progresses:
+
+- A ``started`` message (``role="tool"``) is appended immediately when the
+  tool starts running.
+- An ``intermediate`` message (``role="developer"``) is appended each time an
+  intermediate result is reported via
+  ``result_callback(..., FunctionCallResultProperties(is_final=False))``.
+- A ``final`` message (``role="developer"``) is appended when the task
+  finishes.
+
+This module is the single source of truth for the on-the-wire payload shape:
+
+- The aggregator uses the ``build_*_message`` functions when injecting messages.
+- Realtime LLM services use ``parse_message`` to detect async-tool messages
+  while iterating the context, then read ``payload.result`` and deliver it via
+  their formal tool-result channel.
+
+Internally, ``AsyncToolMessagePayload`` is the canonical structured form;
+the on-the-wire JSON string is always derived from it (never stored) so the
+two representations can't drift.
+
+Consumers are expected to import the module rather than its individual
+functions, e.g.::
+
+    from pipecat.processors.aggregators import async_tool_messages
+    ...
+    async_tool_messages.build_started_message(tool_call_id)
+    async_tool_messages.parse_message(msg)
+"""
+
+import json
+from dataclasses import dataclass
+from typing import Any, Literal
+
+from pipecat.processors.aggregators.llm_context import LLMStandardMessage
+
+AsyncToolMessageKind = Literal["started", "intermediate", "final"]
+
+# --- Payload shape (private; canonical source of truth) ---------------------
+
+# The ``type`` field that identifies an async-tool message payload. Both the
+# builders and the parser use this constant; do not duplicate the literal.
+_PAYLOAD_TYPE = "async_tool"
+
+# Status value for started / intermediate messages (task still running).
+_STATUS_RUNNING = "running"
+
+# Status value for the final message (task complete).
+_STATUS_FINISHED = "finished"
+
+# Description shipped on the started message. The text is intentionally
+# self-explanatory so a model reading the context can tell what's about to
+# happen even without out-of-band knowledge of the protocol.
+_STARTED_DESCRIPTION = (
+    "An asynchronous task associated with this tool_call_id has started "
+    "running. Expect results to arrive later as developer messages that look "
+    "roughly like this one (with 'type=async_tool' and a matching tool_call_id) "
+    "but with a 'result' field. Note that there *may* be more than one result "
+    "(i.e., a stream of results), but there doesn't have to be (there may be "
+    "only one). The last result will come in a message with 'status=finished'."
+)
+
+# Description shipped on each intermediate-result message.
+_INTERMEDIATE_DESCRIPTION = (
+    "This is an intermediate result for the asynchronous task associated with "
+    "this tool_call_id. The task is still running. More intermediate results "
+    "may follow, or the next result may be the final one with "
+    "'status=finished'."
+)
+
+# Description shipped on the final-result message.
+_FINAL_DESCRIPTION = (
+    "This is the final result for the asynchronous task associated with this "
+    "tool_call_id. The task has completed. No further results will arrive for "
+    "this tool_call_id."
+)
+
+
+@dataclass(frozen=True)
+class AsyncToolMessagePayload:
+    """The structured contents of an async-tool message in an LLM context.
+
+    Parameters:
+        kind: Which of the three async-tool message stages this is.
+        tool_call_id: The id of the tool invocation this payload relates to.
+        status: ``"running"`` for started/intermediate, ``"finished"`` for
+            the final message.
+        description: Human-readable description from the payload. May be empty.
+        result: For ``intermediate`` and ``final`` messages, the JSON-encoded
+            result string (or the literal ``"COMPLETED"`` if the function
+            returned no value). ``None`` for ``started`` messages.
+    """
+
+    kind: AsyncToolMessageKind
+    tool_call_id: str
+    status: Literal["running", "finished"]
+    description: str
+    result: str | None
+
+
+# --- Internal: payload ↔ on-the-wire forms -----------------------------------
+
+
+def _payload_to_json(payload: AsyncToolMessagePayload) -> str:
+    """Serialize a payload to its on-the-wire JSON string form.
+
+    Fields that don't apply to the payload's kind are omitted (notably
+    ``result`` is left out of ``started`` payloads, since the task hasn't
+    produced a result yet).
+    """
+    obj: dict[str, Any] = {
+        "type": _PAYLOAD_TYPE,
+        "status": payload.status,
+        "tool_call_id": payload.tool_call_id,
+        "description": payload.description,
+    }
+    if payload.result is not None:
+        obj["result"] = payload.result
+    return json.dumps(obj)
+
+
+def _payload_to_message(payload: AsyncToolMessagePayload) -> LLMStandardMessage:
+    """Wrap a payload in the LLM context message shape that matches its kind.
+
+    - ``started``: ``role="tool"`` plus ``tool_call_id`` at the top level
+      (so the message can sit alongside other regular tool-result messages).
+    - ``intermediate`` / ``final``: ``role="developer"``; ``tool_call_id``
+      lives only inside the JSON payload.
+    """
+    content = _payload_to_json(payload)
+    if payload.kind == "started":
+        return {
+            "role": "tool",
+            "content": content,
+            "tool_call_id": payload.tool_call_id,
+        }
+    return {
+        "role": "developer",
+        "content": content,
+    }
+
+
+# --- Builders ----------------------------------------------------------------
+
+
+def build_started_message(tool_call_id: str) -> LLMStandardMessage:
+    """Build a ``started`` async-tool message for an LLM context.
+
+    Append the returned message to the LLM context immediately when an async
+    function call (registered with ``cancel_on_interruption=False``) starts
+    running. The message lets the model know a task is in flight and that its
+    results will arrive later in subsequent ``developer``-role messages.
+
+    Args:
+        tool_call_id: The id of the tool invocation this message is for.
+
+    Returns:
+        A message ready to pass to ``LLMContext.add_message``.
+    """
+    return _payload_to_message(
+        AsyncToolMessagePayload(
+            kind="started",
+            tool_call_id=tool_call_id,
+            status=_STATUS_RUNNING,
+            description=_STARTED_DESCRIPTION,
+            result=None,
+        )
+    )
+
+
+def build_intermediate_result_message(tool_call_id: str, result: str) -> LLMStandardMessage:
+    """Build an intermediate-result async-tool message for an LLM context.
+
+    Append the returned message to the LLM context each time the running async
+    function reports a non-final result via
+    ``result_callback(..., FunctionCallResultProperties(is_final=False))``.
+
+    Args:
+        tool_call_id: The id of the tool invocation the result is for.
+        result: The JSON-encoded result string (caller is responsible for
+            encoding the function's actual return value, typically via
+            ``json.dumps``).
+
+    Returns:
+        A message ready to pass to ``LLMContext.add_message``.
+    """
+    return _payload_to_message(
+        AsyncToolMessagePayload(
+            kind="intermediate",
+            tool_call_id=tool_call_id,
+            status=_STATUS_RUNNING,
+            description=_INTERMEDIATE_DESCRIPTION,
+            result=result,
+        )
+    )
+
+
+def build_final_result_message(tool_call_id: str, result: str) -> LLMStandardMessage:
+    """Build a final-result async-tool message for an LLM context.
+
+    Append the returned message to the LLM context when the async function
+    finishes. After this message no further async-tool messages will arrive
+    for this ``tool_call_id``.
+
+    Args:
+        tool_call_id: The id of the tool invocation the result is for.
+        result: The JSON-encoded result string, or the literal ``"COMPLETED"``
+            sentinel when the function returned ``None`` (matching the same
+            convention used for synchronous tool calls).
+
+    Returns:
+        A message ready to pass to ``LLMContext.add_message``.
+    """
+    return _payload_to_message(
+        AsyncToolMessagePayload(
+            kind="final",
+            tool_call_id=tool_call_id,
+            status=_STATUS_FINISHED,
+            description=_FINAL_DESCRIPTION,
+            result=result,
+        )
+    )
+
+
+# --- Parsing -----------------------------------------------------------------
+
+
+def parse_message(message: LLMStandardMessage) -> AsyncToolMessagePayload | None:
+    """Decode an async-tool message payload, or return None if not async-tool.
+
+    Args:
+        message: A standard message from the LLM context. Callers iterating
+            over ``LLMContext.get_messages()`` should filter out
+            ``LLMSpecificMessage`` entries first; only ``LLMStandardMessage``
+            values can carry async-tool payloads.
+
+    Returns:
+        An ``AsyncToolMessagePayload`` if the message is a recognized
+        async-tool payload, otherwise ``None``.
+    """
+    role = message.get("role")
+    if role not in ("tool", "developer"):
+        return None
+    content = message.get("content")
+    if not isinstance(content, str):
+        return None
+    try:
+        payload = json.loads(content)
+    except (json.JSONDecodeError, ValueError):
+        return None
+    if not isinstance(payload, dict) or payload.get("type") != _PAYLOAD_TYPE:
+        return None
+    tool_call_id = payload.get("tool_call_id")
+    status = payload.get("status")
+    if not isinstance(tool_call_id, str) or status not in (_STATUS_RUNNING, _STATUS_FINISHED):
+        return None
+    description = payload.get("description", "")
+    if not isinstance(description, str):
+        description = ""
+    result = payload.get("result")
+    if result is not None and not isinstance(result, str):
+        result = None
+    if result is None:
+        kind: AsyncToolMessageKind = "started"
+    elif status == _STATUS_FINISHED:
+        kind = "final"
+    else:
+        kind = "intermediate"
+    return AsyncToolMessagePayload(
+        kind=kind,
+        tool_call_id=tool_call_id,
+        status=status,
+        description=description,
+        result=result,
+    )

src/pipecat/processors/aggregators/llm_context.py

@@ -21,7 +21,7 @@ import io
 import wave
 from collections.abc import Callable
 from dataclasses import dataclass
-from typing import Any, TypeAlias, TypeGuard, TypeVar
+from typing import Any, TypeAlias, TypeGuard, TypeVar, cast
 
 from loguru import logger
 from openai._types import NOT_GIVEN as OPEN_AI_NOT_GIVEN
@@ -129,13 +129,13 @@ class LLMContext:
             url: The URL of the image.
             text: Optional text to include with the image.
         """
-        content = []
+        content: list[dict[str, Any]] = []
         if text:
             content.append({"type": "text", "text": text})
 
         content.append({"type": "image_url", "image_url": {"url": url}})
 
-        return {"role": role, "content": content}
+        return cast(LLMContextMessage, {"role": role, "content": content})
 
     @staticmethod
     async def create_image_message(
@@ -187,7 +187,7 @@ class LLMContext:
             audio_frames: List of audio frame objects to include.
             text: Optional text to include with the audio.
         """
-        content = [{"type": "text", "text": text}]
+        content: list[dict[str, Any]] = [{"type": "text", "text": text}]
 
         def encode_audio():
             sample_rate = audio_frames[0].sample_rate
@@ -214,7 +214,7 @@ class LLMContext:
             }
         )
 
-        return {"role": role, "content": content}
+        return cast(LLMContextMessage, {"role": role, "content": content})
 
     @property
     def messages(self) -> list[LLMContextMessage]:
@@ -295,7 +295,10 @@ class LLMContext:
                 result.append(msg_copy)
                 continue
 
-            msg = copy.deepcopy(message)
+            # The standard message variant is a union of TypedDicts; the
+            # mutations below operate on plain dicts at runtime. Treat as
+            # such for the duration of the redaction loop.
+            msg: dict[str, Any] = cast(dict[str, Any], copy.deepcopy(message))
             content = msg.get("content")
             if isinstance(content, list):
                 for item in content:

src/pipecat/processors/aggregators/llm_response_universal.py

@@ -25,6 +25,7 @@ from pipecat.adapters.schemas.tools_schema import ToolsSchema
 from pipecat.audio.vad.vad_analyzer import VADAnalyzer
 from pipecat.audio.vad.vad_controller import VADController
 from pipecat.frames.frames import (
+    AggregatedTextFrame,
     AssistantImageRawFrame,
     BotStartedSpeakingFrame,
     BotStoppedSpeakingFrame,
@@ -44,6 +45,7 @@ from pipecat.frames.frames import (
     LLMContextSummaryRequestFrame,
     LLMFullResponseEndFrame,
     LLMFullResponseStartFrame,
+    LLMMarkerFrame,
     LLMMessagesAppendFrame,
     LLMMessagesTransformFrame,
     LLMMessagesUpdateFrame,
@@ -53,7 +55,6 @@ from pipecat.frames.frames import (
     LLMThoughtEndFrame,
     LLMThoughtStartFrame,
     LLMThoughtTextFrame,
-    LLMUpdateSettingsFrame,
     StartFrame,
     TextFrame,
     TranscriptionFrame,
@@ -67,25 +68,29 @@ from pipecat.frames.frames import (
     VADUserStartedSpeakingFrame,
     VADUserStoppedSpeakingFrame,
 )
+from pipecat.processors.aggregators import async_tool_messages
 from pipecat.processors.aggregators.llm_context import (
     LLMContext,
     LLMContextMessage,
     LLMSpecificMessage,
     NotGiven,
+    is_given,
 )
 from pipecat.processors.aggregators.llm_context_summarizer import (
     LLMContextSummarizer,
     SummaryAppliedEvent,
 )
 from pipecat.processors.frame_processor import FrameDirection, FrameProcessor
-from pipecat.services.settings import LLMSettings
 from pipecat.turns.user_idle_controller import UserIdleController
 from pipecat.turns.user_mute import BaseUserMuteStrategy
 from pipecat.turns.user_start import BaseUserTurnStartStrategy, UserTurnStartedParams
 from pipecat.turns.user_stop import BaseUserTurnStopStrategy, UserTurnStoppedParams
 from pipecat.turns.user_turn_completion_mixin import UserTurnCompletionConfig
 from pipecat.turns.user_turn_controller import UserTurnController
-from pipecat.turns.user_turn_strategies import UserTurnStrategies
+from pipecat.turns.user_turn_strategies import (
+    FilterIncompleteUserTurnStrategies,
+    UserTurnStrategies,
+)
 from pipecat.utils.context.llm_context_summarization import (
     LLMAutoContextSummarizationConfig,
     LLMContextSummarizationConfig,
@@ -99,6 +104,21 @@ class LLMUserAggregatorParams:
     """Parameters for configuring LLM user aggregation behavior.
 
     Parameters:
+        add_tool_change_messages: When True, on each ``LLMSetToolsFrame`` the
+            aggregator computes the diff against the currently advertised tools
+            and appends a developer-role message to the context describing
+            additions/removals. Helps the LLM stay coherent across
+            mid-conversation tool changes, mitigating several flavors of
+            tool-call-related hallucination: calling tools that have been
+            removed, avoiding tools that have been re-added, and hallucinating
+            output (made-up answers or tool-call-shaped non-tool-calls) when
+            tools are unavailable. Only standard tools are diffed; custom
+            (LLM-specific) tools are ignored. When using
+            ``LLMContextAggregatorPair``, prefer setting this via its
+            ``add_tool_change_messages`` argument instead. Defaults to False.
+        audio_idle_timeout: Timeout in seconds to force speech stop when
+            no audio frames are received while in SPEAKING state (e.g. user mutes
+            mic mid-speech). Set to 0 to disable. Defaults to 1.0.
         user_turn_strategies: User turn start and stop strategies.
         user_mute_strategies: List of user mute strategies.
         user_turn_stop_timeout: Time in seconds to wait before considering the
@@ -108,27 +128,64 @@ class LLMUserAggregatorParams:
             has been idle (not speaking) for this duration. Set to 0 to disable
             idle detection.
         vad_analyzer: Voice Activity Detection analyzer instance.
-        audio_idle_timeout: Timeout in seconds to force speech stop when
-            no audio frames are received while in SPEAKING state (e.g. user mutes
-            mic mid-speech). Set to 0 to disable. Defaults to 1.0.
-        filter_incomplete_user_turns: Whether to filter out incomplete user turns.
-            When enabled, the LLM outputs a turn completion marker at the start of
-            each response: ✓ (complete), ○ (incomplete short), or ◐ (incomplete long).
-            Incomplete responses are suppressed and timeouts trigger re-prompting.
-        user_turn_completion_config: Configuration for turn completion behavior including
-            custom instructions, timeouts, and prompts. Only used when
-            filter_incomplete_user_turns is True.
+        filter_incomplete_user_turns: [DEPRECATED] Use
+            ``user_turn_strategies=FilterIncompleteUserTurnStrategies()``
+            instead. When enabled, the LLM outputs a turn-completion
+            marker at the start of each response: ✓ (complete), ○
+            (incomplete short), or ◐ (incomplete long). Incomplete
+            responses are suppressed and timeouts trigger re-prompting.
+
+            .. deprecated:: 1.2.0
+                Use ``user_turn_strategies=FilterIncompleteUserTurnStrategies()``
+                instead. Will be removed in version 2.0.0.
+
+        user_turn_completion_config: [DEPRECATED] Configuration for turn
+            completion behavior including custom instructions, timeouts, and
+            prompts. Only used when filter_incomplete_user_turns is True
+            (deprecated path) — for the new strategy-based API, pass the config
+            directly to ``FilterIncompleteUserTurnStrategies(config=...)``.
+
+            .. deprecated:: 1.2.0
+                Pass the config directly to
+                ``FilterIncompleteUserTurnStrategies(config=...)`` instead.
+                Will be removed in version 2.0.0.
     """
 
+    add_tool_change_messages: bool = False
+    audio_idle_timeout: float = 1.0
     user_turn_strategies: UserTurnStrategies | None = None
     user_mute_strategies: list[BaseUserMuteStrategy] = field(default_factory=list)
     user_turn_stop_timeout: float = 5.0
     user_idle_timeout: float = 0
     vad_analyzer: VADAnalyzer | None = None
-    audio_idle_timeout: float = 1.0
     filter_incomplete_user_turns: bool = False
     user_turn_completion_config: UserTurnCompletionConfig | None = None
 
+    def __post_init__(self):
+        if self.filter_incomplete_user_turns:
+            warnings.warn(
+                "LLMUserAggregatorParams.filter_incomplete_user_turns is deprecated. "
+                "Use user_turn_strategies=FilterIncompleteUserTurnStrategies() instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
+        if self.user_turn_completion_config:
+            warnings.warn(
+                "LLMUserAggregatorParams.user_turn_completion_config is deprecated. "
+                "Use user_turn_strategies=FilterIncompleteUserTurnStrategies() instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
+
+        if self.user_turn_completion_config is not None:
+            warnings.warn(
+                "LLMUserAggregatorParams.user_turn_completion_config is deprecated. "
+                "Pass the config directly to "
+                "FilterIncompleteUserTurnStrategies(config=...) instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
+
 
 @dataclass
 class LLMAssistantAggregatorParams:
@@ -143,14 +200,32 @@ class LLMAssistantAggregatorParams:
             summarization. Controls trigger thresholds, message preservation, and
             summarization prompts. If None, uses default
             ``LLMAutoContextSummarizationConfig`` values.
+        add_tool_change_messages: When True, on each ``LLMSetToolsFrame`` the
+            aggregator computes the diff against the currently advertised tools
+            and appends a developer-role message to the context describing
+            additions/removals. Helps the LLM stay coherent across
+            mid-conversation tool changes, mitigating several flavors of
+            tool-call-related hallucination: calling tools that have been
+            removed, avoiding tools that have been re-added, and hallucinating
+            output (made-up answers or tool-call-shaped non-tool-calls) when
+            tools are unavailable. Only standard tools are diffed; custom
+            (LLM-specific) tools are ignored. When using
+            ``LLMContextAggregatorPair``, prefer setting this via its
+            ``add_tool_change_messages`` argument instead. Defaults to False.
     """
 
     enable_auto_context_summarization: bool = False
     auto_context_summarization_config: LLMAutoContextSummarizationConfig | None = None
+    add_tool_change_messages: bool = False
 
     # ---------------------------------------------------------------------------
     # Deprecated field names — kept for backward compatibility.
     # Use enable_auto_context_summarization and auto_context_summarization_config instead.
+    #
+    # .. deprecated:: 1.2.0
+    #     Use ``enable_auto_context_summarization`` and
+    #     ``auto_context_summarization_config`` instead. Will be removed in
+    #     version 2.0.0.
     # ---------------------------------------------------------------------------
     enable_context_summarization: bool | None = None
     context_summarization_config: LLMContextSummarizationConfig | None = None
@@ -248,20 +323,87 @@ class LLMContextAggregator(FrameProcessor):
     common functionality for context-based conversation management.
     """
 
-    def __init__(self, *, context: LLMContext, role: str, **kwargs):
+    # Developer-role messages appended to the context when tools are added/
+    # removed via ``LLMSetToolsFrame`` (only when ``add_tool_change_messages``
+    # is enabled on the aggregator's params). ``{function_names}`` is
+    # substituted with a sorted, comma-separated, backtick-wrapped list.
+    TOOL_ACTIVATION_MESSAGE_TEMPLATE = (
+        "The following function(s) have just been added and may now be called: "
+        "{function_names}. Any previously available functions remain available."
+    )
+    TOOL_DEACTIVATION_MESSAGE_TEMPLATE = (
+        "The following function(s) have just been removed and should not be called: "
+        "{function_names}. Any previously available functions remain available. "
+        "The removed function(s) may become available again later, in which case "
+        "you will be informed."
+    )
+
+    def __init__(
+        self,
+        *,
+        context: LLMContext,
+        role: str,
+        add_tool_change_messages: bool = False,
+        **kwargs,
+    ):
         """Initialize the context response aggregator.
 
         Args:
             context: The LLM context to use for conversation storage.
             role: The role this aggregator represents (e.g. "user", "assistant").
+            add_tool_change_messages: See the field of the same name on the
+                aggregator-specific params dataclasses. Subclasses propagate
+                this from their ``params``.
             **kwargs: Additional arguments passed to parent class.
         """
         super().__init__(**kwargs)
         self._context = context
         self._role = role
+        self._add_tool_change_messages = add_tool_change_messages
 
         self._aggregation: list[TextPartForConcatenation] = []
 
+    def _maybe_add_tool_change_messages(self, new_tools: ToolsSchema | NotGiven) -> None:
+        """Append a developer message describing tool add/remove deltas.
+
+        No-op unless ``add_tool_change_messages`` was enabled on the aggregator,
+        and no-op when the diff against the currently advertised tools is empty.
+        Custom (LLM-specific) tools are ignored — only standard tools are diffed.
+
+        Both aggregators call this on every ``LLMSetToolsFrame`` they handle.
+        Whichever aggregator handles the frame first computes a real diff
+        against the shared context and adds the announcement; by the time
+        the other aggregator sees it (if at all), the context already
+        reflects the new tools, so its diff is empty and no duplicate
+        message is added. This is order-independent: it works whether the
+        frame flows downstream (user aggregator first) or upstream
+        (assistant aggregator first, and consumed without being forwarded).
+        """
+        if not self._add_tool_change_messages:
+            return
+
+        def _names(tools: ToolsSchema | NotGiven) -> set[str]:
+            if not is_given(tools):
+                return set()
+            return {s.name for s in tools.standard_tools}
+
+        old_names = _names(self._context.tools)
+        new_names = _names(new_tools)
+        added = new_names - old_names
+        removed = old_names - new_names
+        if not added and not removed:
+            return
+
+        parts: list[str] = []
+        if added:
+            names = ", ".join(f"`{n}`" for n in sorted(added))
+            parts.append(self.TOOL_ACTIVATION_MESSAGE_TEMPLATE.format(function_names=names))
+        if removed:
+            names = ", ".join(f"`{n}`" for n in sorted(removed))
+            parts.append(self.TOOL_DEACTIVATION_MESSAGE_TEMPLATE.format(function_names=names))
+
+        self._context.add_message({"role": "developer", "content": " ".join(parts)})
+
     @property
     def messages(self) -> list[LLMContextMessage]:
         """Get messages from the LLM context.
@@ -434,20 +576,46 @@ class LLMUserAggregator(LLMContextAggregator):
             params: Configuration parameters for aggregation behavior.
             **kwargs: Additional arguments.
         """
-        super().__init__(context=context, role="user", **kwargs)
-        self._params = params or LLMUserAggregatorParams()
+        params = params or LLMUserAggregatorParams()
+        super().__init__(
+            context=context,
+            role="user",
+            add_tool_change_messages=params.add_tool_change_messages,
+            **kwargs,
+        )
+        self._params = params
 
         self._register_event_handler("on_user_turn_started")
         self._register_event_handler("on_user_turn_stopped")
         self._register_event_handler("on_user_turn_stop_timeout")
         self._register_event_handler("on_user_turn_idle")
+        self._register_event_handler("on_user_turn_inference_triggered")
         self._register_event_handler("on_user_mute_started")
         self._register_event_handler("on_user_mute_stopped")
 
         user_turn_strategies = self._params.user_turn_strategies or UserTurnStrategies()
 
+        # Deprecated path: translate filter_incomplete_user_turns into
+        # the equivalent FilterIncompleteUserTurnStrategies wiring. The
+        # DeprecationWarning is emitted in LLMUserAggregatorParams.__post_init__.
+        if self._params.filter_incomplete_user_turns:
+            user_turn_strategies = FilterIncompleteUserTurnStrategies(
+                start=user_turn_strategies.start,
+                stop=user_turn_strategies.stop,
+                config=self._params.user_turn_completion_config,
+            )
+            self._params.user_turn_strategies = user_turn_strategies
+
         self._user_is_muted = False
         self._user_turn_start_timestamp = ""
+        # Full transcript across the user turn. Each
+        # `_on_user_turn_inference_triggered` push captures only the
+        # new segment since the previous push (push_aggregation resets
+        # `_aggregation` after writing to context); we accumulate those
+        # segments here so the eventual `on_user_turn_stopped` event
+        # surfaces the full turn transcript even when several
+        # inferences fire before finalization.
+        self._full_user_turn_aggregation: str | None = None
 
         self._user_turn_controller = UserTurnController(
             user_turn_strategies=user_turn_strategies,
@@ -458,6 +626,9 @@ class LLMUserAggregator(LLMContextAggregator):
         self._user_turn_controller.add_event_handler(
             "on_user_turn_started", self._on_user_turn_started
         )
+        self._user_turn_controller.add_event_handler(
+            "on_user_turn_inference_triggered", self._on_user_turn_inference_triggered
+        )
         self._user_turn_controller.add_event_handler(
             "on_user_turn_stopped", self._on_user_turn_stopped
         )
@@ -536,6 +707,7 @@ class LLMUserAggregator(LLMContextAggregator):
         elif isinstance(frame, LLMMessagesTransformFrame):
             await self._handle_llm_messages_transform(frame)
         elif isinstance(frame, LLMSetToolsFrame):
+            self._maybe_add_tool_change_messages(frame.tools)
             self.set_tools(frame.tools)
             # Push the LLMSetToolsFrame as well, since speech-to-speech LLM
             # services (like OpenAI Realtime) may need to know about tool
@@ -575,21 +747,6 @@ class LLMUserAggregator(LLMContextAggregator):
         for s in self._params.user_mute_strategies:
             await s.setup(self.task_manager)
 
-        # Enable incomplete turn filtering on the LLM if configured
-        if self._params.filter_incomplete_user_turns:
-            # Get config or use defaults
-            config = self._params.user_turn_completion_config or UserTurnCompletionConfig()
-
-            # Enable the feature on the LLM with config
-            await self.push_frame(
-                LLMUpdateSettingsFrame(
-                    delta=LLMSettings(
-                        filter_incomplete_user_turns=True,
-                        user_turn_completion_config=config,
-                    )
-                )
-            )
-
     async def _stop(self, frame: EndFrame):
         await self._maybe_emit_user_turn_stopped(on_session_end=True)
         await self._cleanup()
@@ -729,6 +886,7 @@ class LLMUserAggregator(LLMContextAggregator):
         logger.debug(f"{self}: User started speaking (strategy: {strategy})")
 
         self._user_turn_start_timestamp = time_now_iso8601()
+        self._full_user_turn_aggregation = None
 
         if params.enable_user_speaking_frames:
             await self.broadcast_frame(UserStartedSpeakingFrame)
@@ -740,6 +898,30 @@ class LLMUserAggregator(LLMContextAggregator):
 
         await self._call_event_handler("on_user_turn_started", strategy)
 
+    async def _on_user_turn_inference_triggered(
+        self,
+        controller: UserTurnController,
+        strategy: BaseUserTurnStopStrategy,
+    ):
+        logger.debug(f"{self}: User turn inference triggered (strategy: {strategy})")
+
+        # Push aggregation now: this writes the user message segment to
+        # the context and emits LLMContextFrame, which kicks LLM
+        # inference. Concatenate the segment into
+        # `_full_user_turn_aggregation` so multiple inferences in the
+        # same turn don't lose earlier segments from the eventual
+        # `on_user_turn_stopped` event.
+        segment = await self.push_aggregation()
+        if segment:
+            if self._full_user_turn_aggregation:
+                self._full_user_turn_aggregation = (
+                    f"{self._full_user_turn_aggregation} {segment}".strip()
+                )
+            else:
+                self._full_user_turn_aggregation = segment
+
+        await self._call_event_handler("on_user_turn_inference_triggered", strategy)
+
     async def _on_user_turn_stopped(
         self,
         controller: UserTurnController,
@@ -774,15 +956,29 @@ class LLMUserAggregator(LLMContextAggregator):
     ):
         """Maybe emit user turn stopped event.
 
+        Earlier inference triggers in the same turn have already pushed
+        their segments to the context and accumulated them into
+        ``self._full_user_turn_aggregation``. Any aggregation that
+        arrived after the last inference trigger is flushed here so
+        end-of-turn content is never lost from the public event.
+
         Args:
             strategy: The strategy that triggered the turn stop.
             on_session_end: If True, only emit if there's unemitted content
                 (avoids duplicate events when session ends).
         """
-        aggregation = await self.push_aggregation()
-        if not on_session_end or aggregation:
+        segment = await self.push_aggregation()
+        full_aggregation = self._full_user_turn_aggregation
+        self._full_user_turn_aggregation = None
+
+        if segment and full_aggregation:
+            content = f"{full_aggregation} {segment}".strip()
+        else:
+            content = full_aggregation or segment
+
+        if not on_session_end or content:
             message = UserTurnStoppedMessage(
-                content=aggregation, timestamp=self._user_turn_start_timestamp
+                content=content, timestamp=self._user_turn_start_timestamp
             )
             await self._call_event_handler("on_user_turn_stopped", strategy, message)
             self._user_turn_start_timestamp = ""
@@ -843,8 +1039,14 @@ class LLMAssistantAggregator(LLMContextAggregator):
             params: Configuration parameters for aggregation behavior.
             **kwargs: Additional arguments.
         """
-        super().__init__(context=context, role="assistant", **kwargs)
-        self._params = params or LLMAssistantAggregatorParams()
+        params = params or LLMAssistantAggregatorParams()
+        super().__init__(
+            context=context,
+            role="assistant",
+            add_tool_change_messages=params.add_tool_change_messages,
+            **kwargs,
+        )
+        self._params = params
 
         self._function_calls_in_progress: dict[str, FunctionCallInProgressFrame | None] = {}
         self._function_calls_image_results: dict[str, UserImageRawFrame] = {}
@@ -927,13 +1129,15 @@ class LLMAssistantAggregator(LLMContextAggregator):
             await self._handle_end_or_cancel(frame)
             await self.push_frame(frame, direction)
         elif isinstance(frame, LLMAssistantPushAggregationFrame):
-            await self.push_aggregation()
+            await self._handle_push_aggregation()
         elif isinstance(frame, LLMFullResponseStartFrame):
             await self._handle_llm_start(frame)
         elif isinstance(frame, LLMFullResponseEndFrame):
             await self._handle_llm_end(frame)
         elif isinstance(frame, TextFrame):
             await self._handle_text(frame)
+        elif isinstance(frame, LLMMarkerFrame):
+            await self._handle_marker_frame(frame)
         elif isinstance(frame, LLMThoughtStartFrame):
             await self._handle_thought_start(frame)
         elif isinstance(frame, LLMThoughtTextFrame):
@@ -949,6 +1153,7 @@ class LLMAssistantAggregator(LLMContextAggregator):
         elif isinstance(frame, LLMMessagesTransformFrame):
             await self._handle_llm_messages_transform(frame)
         elif isinstance(frame, LLMSetToolsFrame):
+            self._maybe_add_tool_change_messages(frame.tools)
             self.set_tools(frame.tools)
         elif isinstance(frame, LLMSetToolChoiceFrame):
             self.set_tool_choice(frame.tool_choice)
@@ -1075,23 +1280,7 @@ class LLMAssistantAggregator(LLMContextAggregator):
 
         is_async = not frame.cancel_on_interruption
         if is_async:
-            self._context.add_message(
-                {
-                    "role": "tool",
-                    "content": json.dumps(
-                        {
-                            "type": "async_tool",
-                            "status": "running",
-                            "tool_call_id": frame.tool_call_id,
-                            "description": "An asynchronous task associated with this tool_call_id has started running. "
-                            + "Expect results to arrive later as developer messages that look roughly like this one (with 'type=async_tool' and a matching tool_call_id) but with a 'result' field. "
-                            + "Note that there *may* be more than one result (i.e., a stream of results), but there doesn't have to be (there may be only one). "
-                            + "The last result will come in a message with 'status=finished'.",
-                        }
-                    ),
-                    "tool_call_id": frame.tool_call_id,
-                }
-            )
+            self._context.add_message(async_tool_messages.build_started_message(frame.tool_call_id))
         else:
             self._context.add_message(
                 {
@@ -1204,19 +1393,7 @@ class LLMAssistantAggregator(LLMContextAggregator):
 
         result = json.dumps(frame.result, ensure_ascii=False)
         self._context.add_message(
-            {
-                "role": "developer",
-                "content": json.dumps(
-                    {
-                        "type": "async_tool",
-                        "tool_call_id": frame.tool_call_id,
-                        "status": "running",
-                        "description": "This is an intermediate result for the asynchronous task associated with this tool_call_id. "
-                        + "The task is still running. More intermediate results may follow, or the next result may be the final one with 'status=finished'.",
-                        "result": result,
-                    }
-                ),
-            }
+            async_tool_messages.build_intermediate_result_message(frame.tool_call_id, result)
         )
 
     async def _handle_function_call_finished(
@@ -1237,19 +1414,7 @@ class LLMAssistantAggregator(LLMContextAggregator):
             # notified of the completed result instead of updating the IN_PROGRESS
             # tool message.
             self._context.add_message(
-                {
-                    "role": "developer",
-                    "content": json.dumps(
-                        {
-                            "type": "async_tool",
-                            "tool_call_id": frame.tool_call_id,
-                            "status": "finished",
-                            "description": "This is the final result for the asynchronous task associated with this tool_call_id. "
-                            + "The task has completed. No further results will arrive for this tool_call_id.",
-                            "result": result,
-                        }
-                    ),
-                }
+                async_tool_messages.build_final_result_message(frame.tool_call_id, result)
             )
         else:
             self._update_function_call_result(frame.function_name, frame.tool_call_id, result)
@@ -1309,6 +1474,17 @@ class LLMAssistantAggregator(LLMContextAggregator):
     async def _handle_llm_end(self, _: LLMFullResponseEndFrame):
         await self._trigger_assistant_turn_stopped()
 
+    async def _handle_push_aggregation(self):
+        # LLMAssistantPushAggregationFrame is emitted by TTSService at the end
+        # of a TTSSpeakFrame-driven utterance (no surrounding LLM response
+        # cycle), so no LLMFullResponseStartFrame ever set the turn-start
+        # timestamp. Open a turn now so on_assistant_turn_stopped fires for the
+        # greeting text the same way it did before LLMAssistantPushAggregationFrame
+        # was introduced.
+        if not self._assistant_turn_start_timestamp:
+            await self._trigger_assistant_turn_started()
+        await self._trigger_assistant_turn_stopped()
+
     async def _handle_text(self, frame: TextFrame):
         # Skip TextFrame types not intended to build the assistant context
         if isinstance(frame, (TranscriptionFrame, TranslationFrame, InterimTranscriptionFrame)):
@@ -1321,12 +1497,42 @@ class LLMAssistantAggregator(LLMContextAggregator):
         if len(frame.text) == 0:
             return
 
+        text = (
+            frame.raw_text
+            if isinstance(frame, AggregatedTextFrame) and frame.raw_text
+            else frame.text
+        )
         self._aggregation.append(
             TextPartForConcatenation(
-                frame.text, includes_inter_part_spaces=frame.includes_inter_frame_spaces
+                text, includes_inter_part_spaces=frame.includes_inter_frame_spaces
             )
         )
 
+    async def _handle_marker_frame(self, frame: LLMMarkerFrame):
+        if frame.append_to_context_immediately:
+            # Stand-alone marker: write it to the context now as its
+            # own assistant message. Used when the marker is the entire
+            # assistant turn — e.g. the ○ / ◐ incomplete-turn signals,
+            # where the spoken response is suppressed and the marker
+            # is the only artifact.
+            self._context.add_message({"role": "assistant", "content": frame.marker})
+            await self.push_context_frame()
+            timestamp_frame = LLMContextAssistantTimestampFrame(timestamp=time_now_iso8601())
+            await self.push_frame(timestamp_frame)
+            return
+
+        # Marker is part of an in-progress assistant response. Append
+        # it to the running aggregation so `push_aggregation` writes
+        # marker + text as a single context message — e.g. the ✓
+        # complete-turn signal that prefixes the spoken response,
+        # producing "✓ <response>" in context. Markers are stripped
+        # from the transcript via
+        # `_maybe_strip_turn_completion_markers` so consumers see
+        # clean text.
+        self._aggregation.append(
+            TextPartForConcatenation(frame.marker, includes_inter_part_spaces=False)
+        )
+
     async def _handle_thought_start(self, frame: LLMThoughtStartFrame):
         await self._reset_thought_aggregation()
         self._thought_append_to_context = frame.append_to_context
@@ -1478,6 +1684,7 @@ class LLMContextAggregatorPair:
         *,
         user_params: LLMUserAggregatorParams | None = None,
         assistant_params: LLMAssistantAggregatorParams | None = None,
+        add_tool_change_messages: bool | None = None,
     ):
         """Initialize the LLM context aggregator pair.
 
@@ -1485,9 +1692,22 @@ class LLMContextAggregatorPair:
             context: The context to be managed by the aggregators.
             user_params: Parameters for the user context aggregator.
             assistant_params: Parameters for the assistant context aggregator.
+            add_tool_change_messages: When provided, sets the field of the
+                same name on both ``user_params`` and ``assistant_params``,
+                overriding any value already set on either. This is the
+                preferred way to enable tool-change announcements: it ensures
+                both aggregators participate, which makes the feature robust
+                regardless of which aggregator handles a given
+                ``LLMSetToolsFrame``. The shared context guarantees the
+                announcement is added exactly once (the second aggregator's
+                diff is empty by the time it sees the frame). Leave as
+                ``None`` to respect per-params settings.
         """
         user_params = user_params or LLMUserAggregatorParams()
         assistant_params = assistant_params or LLMAssistantAggregatorParams()
+        if add_tool_change_messages is not None:
+            user_params.add_tool_change_messages = add_tool_change_messages
+            assistant_params.add_tool_change_messages = add_tool_change_messages
         self._user = LLMUserAggregator(context, params=user_params)
         self._assistant = LLMAssistantAggregator(context, params=assistant_params)
 

src/pipecat/processors/aggregators/llm_text_processor.py

@@ -23,6 +23,7 @@ from pipecat.frames.frames import (
 )
 from pipecat.processors.frame_processor import FrameDirection, FrameProcessor
 from pipecat.utils.text.base_text_aggregator import BaseTextAggregator
+from pipecat.utils.text.pattern_pair_aggregator import PatternMatch
 from pipecat.utils.text.simple_text_aggregator import SimpleTextAggregator
 
 
@@ -85,7 +86,11 @@ class LLMTextProcessor(FrameProcessor):
             out_frame = AggregatedTextFrame(
                 text=aggregation.text,
                 aggregated_by=aggregation.type,
+                raw_text=aggregation.full_match
+                if isinstance(aggregation, PatternMatch)
+                else aggregation.text,
             )
+            out_frame.append_to_context = True
             out_frame.skip_tts = in_frame.skip_tts
             await self.push_frame(out_frame)
 
@@ -96,6 +101,9 @@ class LLMTextProcessor(FrameProcessor):
             out_frame = AggregatedTextFrame(
                 text=remaining.text,
                 aggregated_by=remaining.type,
+                raw_text=remaining.full_match
+                if isinstance(remaining, PatternMatch)
+                else remaining.text,
             )
             out_frame.skip_tts = skip_tts
             await self.push_frame(out_frame)

src/pipecat/processors/frame_processor.py

@@ -17,7 +17,7 @@ import asyncio
 import dataclasses
 import traceback
 import warnings
-from collections.abc import Awaitable, Callable, Coroutine
+from collections.abc import Awaitable, Callable
 from dataclasses import dataclass
 from enum import Enum
 from typing import (
@@ -217,9 +217,6 @@ class FrameProcessor(BaseObject):
         # Clock
         self._clock: BaseClock | None = None
 
-        # Task Manager
-        self._task_manager: BaseTaskManager | None = None
-
         # Observer
         self._observer: BaseObserver | None = None
 
@@ -368,20 +365,6 @@ class FrameProcessor(BaseObject):
         """
         return self._report_only_initial_ttfb
 
-    @property
-    def task_manager(self) -> BaseTaskManager:
-        """Get the task manager for this processor.
-
-        Returns:
-            The task manager instance.
-
-        Raises:
-            Exception: If the task manager is not initialized.
-        """
-        if not self._task_manager:
-            raise Exception(f"{self} TaskManager is still not initialized.")
-        return self._task_manager
-
     @property
     def pipeline_task(self) -> PipelineTask | None:
         """Get the :class:`PipelineTask` this processor is running in.
@@ -511,43 +494,14 @@ class FrameProcessor(BaseObject):
         await self.stop_processing_metrics()
         await self.stop_text_aggregation_metrics()
 
-    def create_task(self, coroutine: Coroutine, name: str | None = None) -> asyncio.Task:
-        """Create a new task managed by this processor.
-
-        Args:
-            coroutine: The coroutine to run in the task.
-            name: Optional name for the task.
-
-        Returns:
-            The created asyncio task.
-        """
-        if name:
-            name = f"{self}::{name}"
-        else:
-            name = f"{self}::{coroutine.cr_code.co_name}"
-        return self.task_manager.create_task(coroutine, name)
-
-    async def cancel_task(self, task: asyncio.Task, timeout: float | None = 1.0):
-        """Cancel a task managed by this processor.
-
-        A default timeout if 1 second is used in order to avoid potential
-        freezes caused by certain libraries that swallow
-        `asyncio.CancelledError`.
-
-        Args:
-            task: The task to cancel.
-            timeout: Optional timeout for task cancellation.
-        """
-        await self.task_manager.cancel_task(task, timeout)
-
     async def setup(self, setup: FrameProcessorSetup):
         """Set up the processor with required components.
 
         Args:
             setup: Configuration object containing setup parameters.
         """
+        await super().setup(setup.task_manager)
         self._clock = setup.clock
-        self._task_manager = setup.task_manager
         self._observer = setup.observer
         self._pipeline_task = setup.pipeline_task
 
@@ -555,7 +509,7 @@ class FrameProcessor(BaseObject):
         self.__create_input_task()
 
         if self._metrics is not None:
-            await self._metrics.setup(self._task_manager)
+            await self._metrics.setup(self.task_manager)
 
     async def cleanup(self):
         """Clean up processor resources."""
@@ -877,14 +831,19 @@ class FrameProcessor(BaseObject):
             current_is_uninterruptible = isinstance(
                 self.__process_current_frame, UninterruptibleFrame
             )
-            if current_is_uninterruptible or self.__process_queue.has_uninterruptible:
-                # We don't want to cancel an UninterruptibleFrame (either the
-                # one currently being processed or one waiting in the queue),
-                # so we simply cleanup the queue keeping only
-                # UninterruptibleFrames.
+            if current_is_uninterruptible:
+                # The frame currently being processed is uninterruptible, so we
+                # must not cancel it. Just flush non-uninterruptible frames from
+                # the queue; any uninterruptible ones will be kept and processed
+                # after the current frame finishes.
                 self.__reset_process_queue()
             else:
-                # Cancel and re-create the process task.
+                # Cancel and re-create the process task. Previously this branch
+                # was skipped when the queue contained an uninterruptible frame,
+                # which caused slow non-uninterruptible frames to block
+                # interruptions. Uninterruptible queued frames are safe here
+                # because __create_process_task calls __reset_process_queue
+                # internally, which always preserves them.
                 await self.__cancel_process_task()
                 self.__create_process_task()
         except Exception as e: