Update changelog for version 0.0.98

CHANGELOG.md

@@ -7,6 +7,240 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 <!-- towncrier release notes start -->
 
+## [0.0.98] - 2025-12-17
+
+### Added
+
+- Added `RimeNonJsonTTSService` which supports non-JSON streaming mode. This
+  new class supports websocket streaming for the Arcana model.
+  (PR [#3085](https://github.com/pipecat-ai/pipecat/pull/3085))
+
+- Added additional functionality related to "thinking", for Google and
+  Anthropic LLMs.
+
+    1. New typed parameters for Google and Anthropic LLMs that control the
+       models' thinking behavior (like how much thinking to do, and whether to
+       output thoughts or thought summaries):
+       - `AnthropicLLMService.ThinkingConfig`
+       - `GoogleLLMService.ThinkingConfig`
+    2. New frames for representing thoughts output by LLMs:
+       - `LLMThoughtStartFrame`
+       - `LLMThoughtTextFrame`
+       - `LLMThoughtEndFrame`
+    3. A generic mechanism for recording LLM thoughts to context, used
+       specifically to support Anthropic, whose thought signatures are expected
+       to appear alongside the text of the thoughts within assistant context
+       messages. See:
+       - `LLMThoughtEndFrame.signature`
+       - `LLMAssistantAggregator` handling of the above field
+       - `AnthropicLLMAdapter` handling of `"thought"` context messages
+    4. Google-specific logic for inserting thought signatures into the context,
+       to help maintain thinking continuity in a chain of LLM calls. See:
+       - `GoogleLLMService` sending `LLMMessagesAppendFrame`s to add
+         LLM-specific
+         `"thought_signature"` messages to context
+       - `GeminiLLMAdapter` handling of `"thought_signature"` messages
+    5. An expansion of `TranscriptProcessor` to process LLM thoughts in
+       addition to user and assistant utterances. See:
+       - `TranscriptProcessor(process_thoughts=True)` (defaults to `False`)
+       - `ThoughtTranscriptionMessage`, which is now also emitted with the
+         `"on_transcript_update"` event
+  (PR [#3175](https://github.com/pipecat-ai/pipecat/pull/3175))
+
+- Data and control frames can now be marked as non-interruptible by using the
+ `UninterruptibleFrame` mixin. Frames marked as `UninterruptibleFrame` will
+  not be interrupted during processing, and any queued frames of this type will
+  be retained in the internal queues. This is useful when you need ordered
+  frames (data or control) that should not be discarded or cancelled due to
+  interruptions.
+  (PR [#3189](https://github.com/pipecat-ai/pipecat/pull/3189))
+
+- Added `on_conversation_detected` event to `VoicemaiDetector`.
+  (PR [#3207](https://github.com/pipecat-ai/pipecat/pull/3207))
+
+- Added `x-goog-api-client` header with Pipecat's version to all Google
+  services' requests.
+  (PR [#3208](https://github.com/pipecat-ai/pipecat/pull/3208))
+
+- Added support for the HeyGen LiveAvatar API (see https://www.liveavatar.com/).
+  (PR [#3210](https://github.com/pipecat-ai/pipecat/pull/3210))
+
+- Added to `AWSNovaSonicLLMService` functionality related to the new (and now
+  default) Nova 2 Sonic model (`"amazon.nova-2-sonic-v1:0"`):
+    - Added the `endpointing_sensitivity` parameter to control how quickly the
+      model decides the user has stopped speaking.
+    - Made the assistant-response-trigger hack a no-op. It's only needed for
+      the older Nova Sonic model.
+  (PR [#3212](https://github.com/pipecat-ai/pipecat/pull/3212))
+
+- [Ultravox Realtime](https://docs.ultravox.ai) is now a supported
+  speech-to-speech service.
+    - Added `UltravoxRealtimeLLMService` for the integration.
+    - Added `49-ultravox-realtime.py` example (with tool calling).
+  (PR [#3227](https://github.com/pipecat-ai/pipecat/pull/3227))
+
+- Added Daily PSTN dial-in support to the development runner with `--dialin`
+  flag. This includes:
+
+    - `/daily-dialin-webhook` endpoint that handles incoming Daily PSTN webhooks
+    - Automatic Daily room creation with SIP configuration
+    - `DialinSettings` and `DailyDialinRequest` types in `pipecat.runner.types`
+      for type-safe dial-in data
+    - The runner now mimics Pipecat Cloud's dial-in webhook handling for local
+      development
+  (PR [#3235](https://github.com/pipecat-ai/pipecat/pull/3235))
+
+- Add Gladia session id to logs for `GladiaSTTService`.
+  (PR [#3236](https://github.com/pipecat-ai/pipecat/pull/3236))
+
+- Added `InworldHttpTTSService` which uses Inworld's HTTP based TTS service in
+  either streaming or non-streaming mode. Note: This class was previously named
+  `InworldTTSService`.
+  (PR [#3239](https://github.com/pipecat-ai/pipecat/pull/3239))
+
+- Added `language_hints_strict` parameter to `SonioxSTTService` to strictly
+  enforces language hints. This ensures that transcription occurs in the
+  specified language.
+  (PR [#3245](https://github.com/pipecat-ai/pipecat/pull/3245))
+
+- Added Pipecat library version info to the `about` field in the `bot-ready`
+  RTVI message.
+  (PR [#3248](https://github.com/pipecat-ai/pipecat/pull/3248))
+
+- Added `VisionFullResponseStartFrame`, `VisionFullResponseEndFrame` and
+  `VisionTextFrame`. This are used by vision services similar to LLM
+  services.
+  (PR [#3252](https://github.com/pipecat-ai/pipecat/pull/3252))
+
+### Changed
+
+- `FunctionCallInProgressFrame` and `FunctionCallResultFrame` have changed from
+  system frames to a control frame and a data frame, respectively, and are
+  now both marked as `UninterruptibleFrame`.
+  (PR [#3189](https://github.com/pipecat-ai/pipecat/pull/3189))
+
+- `UserBotLatencyLogObserver` now uses `VADUserStartedSpeakingFrame` and
+  `VADUserStoppedSpeakingFrame` to determine latency from user stopped speaking
+  to bot started speaking.
+  (PR [#3206](https://github.com/pipecat-ai/pipecat/pull/3206))
+
+- Updated `HeyGenVideoService` and `HeyGenTransport` to support both HeyGen
+  APIs (Interactive Avatar and Live Avatar).
+  Using them is as simple as specifying the `service_type` when creating the
+  `HeyGenVideoService` and the `HeyGenTransport`:
+    ```python
+    heyGen = HeyGenVideoService(
+        api_key=os.getenv("HEYGEN_LIVE_AVATAR_API_KEY"),
+        service_type=ServiceType.LIVE_AVATAR,
+        session=session,
+    )
+    ```
+  (PR [#3210](https://github.com/pipecat-ai/pipecat/pull/3210))
+
+- Made `"amazon.nova-2-sonic-v1:0"` the new default model for
+  `AWSNovaSonicLLMService`.
+  (PR [#3212](https://github.com/pipecat-ai/pipecat/pull/3212))
+
+- Updated the `run_inference` methods in the LLM service classes
+  (`AnthropicLLMService`, `AWSBedrockLLMService`, `GoogleLLMService`, and
+  `OpenAILLMService` and its base classes) to use the provided LLM
+  configuration parameters.
+  (PR [#3214](https://github.com/pipecat-ai/pipecat/pull/3214))
+
+- Updated default models for:
+    - `GeminiLiveLLMService` to `gemini-2.5-flash-native-audio-preview-12-2025`.
+    - `GeminiLiveVertexLLMService` to `gemini-live-2.5-flash-native-audio`.
+  (PR [#3228](https://github.com/pipecat-ai/pipecat/pull/3228))
+
+- Changed the `reason` field in `EndFrame`, `CancelFrame`, `EndTaskFrame`, and
+  `CancelTaskFrame` from `str` to `Any` to indicate that it can hold values
+  other than strings.
+  (PR [#3231](https://github.com/pipecat-ai/pipecat/pull/3231))
+
+- Updated websocket STT services to use the `WebsocketSTTService` base class.
+  This base class manages the websocket connection and handles reconnects.
+  Updated services:
+    - `AssemblyAISTTService`
+    - `AWSTranscribeSTTService`
+    - `GladiaSTTService`
+    - `SonioxSTTService`
+  (PR [#3236](https://github.com/pipecat-ai/pipecat/pull/3236))
+
+- Changed Inworld's TTS service implementations:
+    - Previously, the HTTP implementation was named `InworldTTSService`. That
+      has been moved to `InworldHttpTTSService`. This service now supports
+      word-timestamp alignment data in both streaming and non-streaming modes.
+    - Updated the `InworldTTSService` class to use Inworld's Websocket API.
+      This class now has support for word-timestamp alignment data and tracks
+      contexts for each user turn.
+  (PR [#3239](https://github.com/pipecat-ai/pipecat/pull/3239))
+
+- ⚠️ Breaking change: `WordTTSService.start_word_timestamps()` and
+    `WordTTSService.reset_word_timestamps()` are now async.
+  (PR [#3240](https://github.com/pipecat-ai/pipecat/pull/3240))
+
+- Updated the current RTVI version to 1.1.0 to reflect recent additions and
+  deprecations.
+    - New RTVI Messages: `send-text` and `bot-output`
+    - Deprecated Messages: `append-to-context` and `bot-transcription`
+  (PR [#3248](https://github.com/pipecat-ai/pipecat/pull/3248))
+
+- `MoondreamService` now pushes `VisionFullResponseStartFrame`,
+   `VisionFullResponseEndFrame` and `VisionTextFrame`.
+  (PR [#3252](https://github.com/pipecat-ai/pipecat/pull/3252))
+
+### Deprecated
+
+- `FalSmartTurnAnalyzer` and `LocalSmartTurnAnalyzer` are deprecated and will
+  be removed in a future version. Use `LocalSmartTurnAnalyzerV3` instead.
+  (PR [#3219](https://github.com/pipecat-ai/pipecat/pull/3219))
+
+### Removed
+
+- Removed the deprecated VLLM-based open source Ultravox STT service.
+  (PR [#3227](https://github.com/pipecat-ai/pipecat/pull/3227))
+
+### Fixed
+
+- Fixed a bug in `AWSNovaSonicLLMService` where we would mishandle cancelled
+  tool calls in the context, resulting in errors.
+  (PR [#3212](https://github.com/pipecat-ai/pipecat/pull/3212))
+
+- Better support conversation history with Gemini 2.5 Flash Image (model
+  "gemini-2.5-flash-image"). Prior to this fix, the model had no memory of
+  previous images it had generated, so it wouldn't be able to iterate on
+  them.
+  (PR [#3224](https://github.com/pipecat-ai/pipecat/pull/3224))
+
+- Support conversations with Gemini 3 Pro Image (model
+  "gemini-3-pro-image-preview"). Prior to this fix, after the model generated
+  an image the conversation would not be able to progress.
+  (PR [#3224](https://github.com/pipecat-ai/pipecat/pull/3224))
+
+- Fixed an issue where `ElevenLabsHttpTTSService` was not updating
+  voice settings when receiving a `TTSUpdateSettingsFrame`.
+  (PR [#3226](https://github.com/pipecat-ai/pipecat/pull/3226))
+
+- Fixed the return type for `SmallWebRTCRequestHandler.handle_web_request()`
+  function.
+  (PR [#3230](https://github.com/pipecat-ai/pipecat/pull/3230))
+
+- Fix a bug in LLM context audio content handling
+  (PR [#3234](https://github.com/pipecat-ai/pipecat/pull/3234))
+
+- In `GladiaSTTService`, reset the `_bytes_sent` counter on connecting the
+  websocket. This avoids unnecessary audio buffer trimming.
+  (PR [#3236](https://github.com/pipecat-ai/pipecat/pull/3236))
+
+- Fixed a TTS service word-timestamp issue that could cause generated
+  `TTSTextFrame` instances to have an incorrect pts (`pts = -1`).
+  (PR [#3240](https://github.com/pipecat-ai/pipecat/pull/3240))
+
+- Fixed an issue in `SimpleTextAggreagtor` where spaces were not being stripped
+  before returning the aggregation. This resulted in an extra space for TTS
+  services that don't support word-timestamp alignment data.
+  (PR [#3247](https://github.com/pipecat-ai/pipecat/pull/3247))
+
 ## [0.0.97] - 2025-12-05
 
 ### Added

changelog/3085.added.md

@@ -1,2 +0,0 @@
-- Added `RimeNonJsonTTSService` which supports non-JSON streaming mode. This new class supports websocket streaming for the Arcana model.
-

changelog/3175.added.md

@@ -1,29 +0,0 @@
-- Added additional functionality related to "thinking", for Google and Anthropic
-  LLMs.
-
-  1. New typed parameters for Google and Anthropic LLMs that control the
-     models' thinking behavior (like how much thinking to do, and whether to
-     output thoughts or thought summaries):
-     - `AnthropicLLMService.ThinkingConfig`
-     - `GoogleLLMService.ThinkingConfig`
-  2. New frames for representing thoughts output by LLMs:
-     - `LLMThoughtStartFrame`
-     - `LLMThoughtTextFrame`
-     - `LLMThoughtEndFrame`
-  3. A generic mechanism for recording LLM thoughts to context, used
-     specifically to support Anthropic, whose thought signatures are expected to
-     appear alongside the text of the thoughts within assistant context
-     messages. See:
-     - `LLMThoughtEndFrame.signature`
-     - `LLMAssistantAggregator` handling of the above field
-     - `AnthropicLLMAdapter` handling of `"thought"` context messages
-  4. Google-specific logic for inserting thought signatures into the context,
-     to help maintain thinking continuity in a chain of LLM calls. See:
-     - `GoogleLLMService` sending `LLMMessagesAppendFrame`s to add LLM-specific
-       `"thought_signature"` messages to context
-     - `GeminiLLMAdapter` handling of `"thought_signature"` messages
-  5. An expansion of `TranscriptProcessor` to process LLM thoughts in addition
-     to user and assistant utterances. See:
-     - `TranscriptProcessor(process_thoughts=True)` (defaults to `False`)
-     - `ThoughtTranscriptionMessage`, which is now also emitted with the
-       `"on_transcript_update"` event

changelog/3189.added.md

@@ -1,6 +0,0 @@
-- Data and control frames can now be marked as non-interruptible by using the
-  `UninterruptibleFrame` mixin. Frames marked as `UninterruptibleFrame` will not
-  be interrupted during processing, and any queued frames of this type will be
-  retained in the internal queues. This is useful when you need ordered frames
-  (data or control) that should not be discarded or cancelled due to
-  interruptions.

changelog/3189.changed.md

@@ -1,3 +0,0 @@
-- `FunctionCallInProgressFrame` and `FunctionCallResultFrame` have changed from
-  system frames to a control frame and a data frame, respectively, and are now
-  both marked as `UninterruptibleFrame`.

changelog/3206.changed.md

@@ -1,3 +0,0 @@
-- `UserBotLatencyLogObserver` now uses `VADUserStartedSpeakingFrame` and 
-`VADUserStoppedSpeakingFrame` to determine latency from user stopped speaking
-to bot started speaking.

changelog/3207.added.md

@@ -1 +0,0 @@
-- Added `on_conversation_detected` event to `VoicemaiDetector`.

changelog/3208.added.md

@@ -1 +0,0 @@
-- Added `x-goog-api-client` header with Pipecat's version to all Google services' requests.

changelog/3210.added.md

@@ -1,2 +0,0 @@
-- Added support for the HeyGen LiveAvatar API
-  (see https://www.liveavatar.com/).

changelog/3210.changed.md

@@ -1,10 +0,0 @@
-- Updated `HeyGenVideoService` and `HeyGenTransport` to support both HeyGen APIs (Interactive Avatar and Live Avatar).
-  Using them is as simple as specifying the `service_type` when creating the `HeyGenVideoService` and the `HeyGenTransport`:
-  ```python
-  heyGen = HeyGenVideoService(
-      api_key=os.getenv("HEYGEN_LIVE_AVATAR_API_KEY"),
-      service_type=ServiceType.LIVE_AVATAR,
-      session=session,
-  )
-  ```
-

changelog/3212.added.md

@@ -1,6 +0,0 @@
-- Added to `AWSNovaSonicLLMService` functionality related to the new (and now
-  default) Nova 2 Sonic model (`"amazon.nova-2-sonic-v1:0"`):
-  - Added the `endpointing_sensitivity` parameter to control how quickly the
-    model decides the user has stopped speaking.
-  - Made the assistant-response-trigger hack a no-op. It's only needed for the
-    older Nova Sonic model.

changelog/3212.changed.md

@@ -1 +0,0 @@
-- Made `"amazon.nova-2-sonic-v1:0"` the new default model for `AWSNovaSonicLLMService`.

changelog/3212.fixed.md

@@ -1,2 +0,0 @@
-- Fixed a bug in `AWSNovaSonicLLMService` where we would mishandle cancelled
-  tool calls in the context, resulting in errors.

changelog/3214.changed.md

@@ -1 +0,0 @@
-- Updated the `run_inference` methods in the LLM service classes (`AnthropicLLMService`, `AWSBedrockLLMService`, `GoogleLLMService`, and `OpenAILLMService` and its base classes) to use the provided LLM configuration parameters.

changelog/3219.deprecated.md

@@ -1 +0,0 @@
-- `FalSmartTurnAnalyzer` and `LocalSmartTurnAnalyzer` are deprecated and will be removed in a future version. Use `LocalSmartTurnAnalyzerV3` instead.

changelog/3224.fixed.2.md

@@ -1,3 +0,0 @@
-- Better support conversation history with Gemini 2.5 Flash Image (model
-  "gemini-2.5-flash-image"). Prior to this fix, the model had no memory of
-  previous images it had generated, so it wouldn't be able to iterate on them.

changelog/3224.fixed.md

@@ -1,3 +0,0 @@
-- Support conversations with Gemini 3 Pro Image (model
-  "gemini-3-pro-image-preview"). Prior to this fix, after the model generated
-  an image the conversation would not be able to progress.

changelog/3226.fixed.md

@@ -1,2 +0,0 @@
-- Fixed an issue where `ElevenLabsHttpTTSService` was not updating 
-  voice settings when receiving a `TTSUpdateSettingsFrame`.

changelog/3227.added.md

@@ -1,4 +0,0 @@
-- [Ultravox Realtime](https://docs.ultravox.ai) is now a supported speech-to-speech
-  service.
-  - Added `UltravoxRealtimeLLMService` for the integration.
-  - Added `49-ultravox-realtime.py` example (with tool calling).

changelog/3227.removed.md

@@ -1 +0,0 @@
-- Removed the deprecated VLLM-based open source Ultravox STT service.

changelog/3228.changed.md

@@ -1,4 +0,0 @@
-- Updated default models for:
-
-  - `GeminiLiveLLMService` to `gemini-2.5-flash-native-audio-preview-12-2025`.
-  - `GeminiLiveVertexLLMService` to `gemini-live-2.5-flash-native-audio`.

changelog/3230.fixed.md

@@ -1 +0,0 @@
-- Fixed the return type for `SmallWebRTCRequestHandler.handle_web_request()` function.

changelog/3231.changed.md

@@ -1,3 +0,0 @@
-- Changed the `reason` field in `EndFrame`, `CancelFrame`, `EndTaskFrame`, and
-  `CancelTaskFrame` from `str` to `Any` to indicate that it can hold values
-  other than strings.

changelog/3234.fixed.md

@@ -1 +0,0 @@
-- Fix a bug in LLM context audio content handling

changelog/3235.added.md

@@ -1,6 +0,0 @@
-- Added Daily PSTN dial-in support to the development runner with `--dialin` flag. This includes:
-
-  - `/daily-dialin-webhook` endpoint that handles incoming Daily PSTN webhooks
-  - Automatic Daily room creation with SIP configuration
-  - `DialinSettings` and `DailyDialinRequest` types in `pipecat.runner.types` for type-safe dial-in data
-  - The runner now mimics Pipecat Cloud's dial-in webhook handling for local development

changelog/3236.added.md

@@ -1 +0,0 @@
-- Add Gladia session id to logs for `GladiaSTTService`.

changelog/3236.changed.md

@@ -1,6 +0,0 @@
-- Updated websocket STT services to use the `WebsocketSTTService` base class. This base class manages the websocket connection and handles reconnects. Updated services:
-
-  - `AssemblyAISTTService`
-  - `AWSTranscribeSTTService`
-  - `GladiaSTTService`
-  - `SonioxSTTService`

changelog/3236.fixed.md

@@ -1 +0,0 @@
-- In `GladiaSTTService`, reset the `_bytes_sent` counter on connecting the websocket. This avoids unnecessary audio buffer trimming.

changelog/3239.added.md

@@ -1 +0,0 @@
-- Added `InworldHttpTTSService` which uses Inworld's HTTP based TTS service in either streaming or non-streaming mode. Note: This class was previously named `InworldTTSService`.

changelog/3239.changed.md

@@ -1,8 +0,0 @@
-- Changed Inworld's TTS service implementations:
-
-  - Previously, the HTTP implementation was named `InworldTTSService`. That has
-    been moved to `InworldHttpTTSService`. This service now supports
-    word-timestamp alignment data in both streaming and non-streaming modes.
-  - Updated the `InworldTTSService` class to use Inworld's Websocket API. This
-    class now has support for word-timestamp alignment data and tracks contexts
-    for each user turn.

changelog/3240.changed.md

@@ -1,2 +0,0 @@
-- ⚠️ Breaking change: `WordTTSService.start_word_timestamps()` and
-  `WordTTSService.reset_word_timestamps()` are now async.

changelog/3240.fixed.md

@@ -1,2 +0,0 @@
-- Fixed a TTS service word-timestamp issue that could cause generated
-  `TTSTextFrame` instances to have an incorrect pts (`pts = -1`).

changelog/3245.added.md

@@ -1 +0,0 @@
-- Added `language_hints_strict` parameter to `SonioxSTTService` to strictly enforces language hints. This ensures that transcription occurs in the specified language.

changelog/3247.fixed.md

@@ -1 +0,0 @@
-- Fixed an issue in `SimpleTextAggreagtor` where spaces were not being stripped before returning the aggregation. This resulted in an extra space for TTS services that don't support word-timestamp alignment data.

changelog/3248.added.md

@@ -1 +0,0 @@
-- Added Pipecat library version info to the `about` field in the `bot-ready` RTVI message.

changelog/3248.changed.md

@@ -1,3 +0,0 @@
-- Updated the current RTVI version to 1.1.0 to reflect recent additions and deprecations.
-  - New RTVI Messages: `send-text` and `bot-output`
-  - Deprecated Messages: `append-to-context` and `bot-transcription`

changelog/3252.added.md

@@ -1,2 +0,0 @@
-- Added `VisionFullResponseStartFrame`, `VisionFullResponseEndFrame` and
-  `VisionTextFrame`. This are used by vision services similar to LLM services.

changelog/3252.changed.md

@@ -1,2 +0,0 @@
-- `MoondreamService` now pushes `VisionFullResponseStartFrame`,
-  `VisionFullResponseEndFrame` and `VisionTextFrame`.