Merge remote-tracking branch 'origin/main' into livekit-transport-helper

# Conflicts:
#	CHANGELOG.md
#	uv.lock

CHANGELOG.md

@@ -11,14 +11,248 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - Added `LiveKitRESTHelper` utility class for managing LiveKit rooms via REST API.
 
+- Added `DeepgramSageMakerSTTService` which connects to a SageMaker hosted
+  Deepgram STT model. Added `07c-interruptible-deepgram-sagemaker.py`
+  foundational example.
+
+- Added `SageMakerBidiClient` to connect to SageMaker hosted BiDi compatible
+  services.
+
+- Added support for `include_timestamps` and `enable_logging` in
+  `ElevenLabsRealtimeSTTService`. When `include_timestamps` is enabled,
+  timestamp data is included in the `TranscriptionFrame`'s `result`
+  parameter.
+
 - Added optional speaking rate control to `InworldTTSService`.
 
+- Introduced a new `AggregatedTextFrame` type to support passing text along with
+  an `aggregated_by` field to describe the type of text
+  included. `TTSTextFrame`s now inherit from `AggregatedTextFrame`. With this
+  inheritance, an observer can watch for `AggregatedTextFrame`s to accumlate the
+  perceived output and determine whether or not the text was spoken based on if
+  that frame is also a `TTSTextFrame`.
+
+  With this frame, the llm token stream can be transformed into custom
+  composable chunks, allowing for aggregation outside the TTS service. This
+  makes it possible to listen for or handle those aggregations and sets the
+  stage for doing things like composing a best effort of the perceived llm
+  output in a more digestable form and to do so whether or not it is processed
+  by a TTS or if even a TTS exists.
+
+- Introduced `LLMTextProcessor`: A new processor meant to allow customization
+  for how LLMTextFrames should be aggregated and considered. It's purpose is to
+  turn `LLMTextFrame`s into `AggregatedTextFrame`s. By default, a TTSService
+  will still aggregate `LLMTextFrame`s by sentence for the service to
+  consume. However, if you wish to override how the llm text is aggregated, you
+  should no longer override the TTS's internal text_aggregator, but instead,
+  insert this processor between your LLM and TTS in the pipeline.
+
+- New `bot-output` RTVI message to represent what the bot actually "says".
+
+  - The `RTVIObserver` now emits `bot-output` messages based off the new
+    `AggregatedTextFrame`s (`bot-tts-text` and `bot-llm-text` are still
+    supported and generated, but `bot-transcript` is now deprecated in lieu of
+    this new, more thorough, message).
+
+  - The new `RTVIBotOutputMessage` includes the fields:
+
+    - `spoken`: A boolean indicating whether the text was spoken by TTS
+
+    - `aggregated_by`: A string representing how the text was aggregated
+      ("sentence", "word", "my custom aggregation")
+
+  - Introduced new fields to `RTVIObserver` to support the new `bot-output`
+    messaging:
+
+    - `bot_output_enabled`: Defaults to True. Set to false to disable bot-output
+      messages.
+
+    - `skip_aggregator_types`: Defaults to `None`. Set to a list of strings that
+      match aggregation types that should not be included in bot-output
+      messages. (Ex. `credit_card`)
+
+  - Introduced new methods, `add_text_transformer()` and
+    `remove_text_transformer()`, to `RTVIObserver` to support providing (and
+    subsequently removing) callbacks for various types of aggregations (or all
+    aggregations with `*`) that can modify the text before being sent as a
+    `bot-output` or `tts-text` message. (Think obscuring the credit card or
+    inserting extra detail the client might want that the context doesn't need.)
+
+- In `MiniMaxHttpTTSService`:
+
+  - Added support for speech-2.6-hd and speech-2.6-turbo models
+
+  - Added languages: Afrikaans, Bulgarian, Catalan, Danish, Persian, Filipino,
+    Hebrew, Croatian, Hungarian, Malay, Norwegian, Nynorsk, Slovak, Slovenian,
+    Swedish, and Tamil
+
+  - Added new emotions: calm and fluent
+
 ### Changed
 
 - Updated `daily-python` to 0.22.0.
 
+- `BaseTextAggregator` changes:
+
+  Modified the BaseTextAggregator type so that when text gets aggregated,
+  metadata can be associated with it. Currently, that just means a `type`, so
+  that the aggregation can be classified or described. Changes made to support
+  this:
+
+  - ⚠️ IMPORTANT: Aggregators are now expected to strip leading/trailing white
+    space characters before returning their aggregation from `aggregation()` or
+    `.text`. This way all aggregators have a consistent contract allowing
+    downstream use to know how to stitch aggregations back together.
+
+  - Introduced a new `Aggregation` dataclass to represent both the aggregated
+    `text` and a string identifying the `type` of aggregation (ex. "sentence",
+    "word", "my custom aggregation")
+
+  - ⚠️ Breaking change: `BaseTextAggregator.text` now returns an `Aggregation`
+    (instead of `str`).
+
+    Before:
+
+    ```python
+    aggregated_text = myAggregator.text
+    ```
+
+    Now:
+
+    ```python
+    aggregated_text = myAggregator.text.text
+    ```
+
+  - ⚠️ Breaking change: `BaseTextAggregator.aggregate()` now returns
+    `Optional[Aggregation]` (instead of `Optional[str]`).
+
+    Before:
+
+    ```python
+    aggregation = myAggregator.aggregate(text)
+    print(f"successfully aggregated text: {aggregation}")
+    ```
+
+    Now:
+
+    ```python
+    aggregation = myAggregator.aggregate(text)
+    if aggregation:
+      print(f"successfully aggregated text: {aggregation.text}")
+    ```
+
+  - `SimpleTextAggregator`, `SkipTagsAggregator`, `PatternPairAggregator`
+    updated to produce/consume `Aggregation` objects.
+
+  - All uses of the above Aggregators have been updated accordingly.
+
+- Augmented the `PatternPairAggregator` so that matched patterns can be treated
+  as their own aggregation, taking advantage of the new. To that end:
+
+  - Introduced a new, preferred version of `add_pattern` to support a new option
+    for treating a match as a separate aggregation returned from
+    `aggregate()`. This replaces the now deprecated `add_pattern_pair` method
+    and you provide a `MatchAction` in lieu of the `remove_match` field.
+
+    - `MatchAction` enum: `REMOVE`, `KEEP`, `AGGREGATE`, allowing customization
+      for how a match should be handled.
+
+      - `REMOVE`: The text along with its delimiters will be removed from the
+        streaming text. Sentence aggregation will continue on as if this text
+        did not exist.
+
+      - `KEEP`: The delimiters will be removed, but the content between them
+        will be kept. Sentence aggregation will continue on with the internal
+        text included.
+
+      - `AGGREGATE`: The delimiters will be removed and the content between will
+        be treated as a separate aggregation. Any text before the start of the
+        pattern will be returned early, whether or not a complete sentence was
+        found. Then the pattern will be returned. Then the aggregation will
+        continue on sentence matching after the closing delimiter is found. The
+        content between the delimiters is not aggregated by sentence. It is
+        aggregated as one single block of text.
+
+    - `PatternMatch` now extends `Aggregation` and provides richer info to
+      handlers.
+
+  - ⚠️ Breaking change: The `PatternMatch` type returned to handlers registered
+    via `on_pattern_match` has been updated to subclass from the new
+    `Aggregation` type, which means that `content` has been replaced with
+    `text` and `pattern_id` has been replaced with `type`:
+
+    ```python
+    async dev on_match_tag(match: PatternMatch):
+       pattern = match.type # instead of match.pattern_id
+       text = match.text # instead of match.content
+    ```
+
+- `TextFrame` now includes the field `append_to_context` to support setting
+  whether or not the encompassing text should be added to the LLM context (by
+  the LLM assistant aggregator). It defaults to `True`.
+
+- `TTSService` base class updates:
+
+  - `TTSService`s now accept a new `skip_aggregator_types` to avoid speaking
+    certain aggregation types (now determined/returned by the aggregator)
+
+  - Introduced the ability to do a just-in-time transform of text before it gets
+    sent to the TTS service via callbacks you can set up via a new init field,
+    `text_transforms` or a new method `add_text_transformer()`. This makes it
+    possible to do things like introduce TTS-specific tags for spelling or
+    emotion or change the pronunciation of something on the
+    fly. `remove_text_transformer` has also been added to support removing a
+    registered transform callback.
+
+  - TTS services push `AggregatedTextFrame` in addition to `TTSTextFrame`s when
+    either an aggregation occurs that should not be spoken or when the TTS
+    service supports word-by-word timestamping. In the latter case, the
+    `TTSService` preliminarily generates an `AggregatedTextFrame`, aggregated by
+    sentence to generate the full sentence content as early as possible.
+
+- Updated `CartesiaTTSService`:
+
+  - Modified use of custom default text_aggregator to avoid deprecation warnings
+    and push users towards use of transformers or the `LLMTextProcessor`
+
+  - Added convenience methods for taking advantage of Cartesia's SSML tags:
+    spell, emotion, pauses, volume, and speed.
+
+- Updated `RimeTTSService`:
+
+  - Modified use of custom default text_aggregator to avoid deprecation warnings
+    and push users towards use of transformers or the `LLMTextProcessor`
+
+  - Added convenience methods for taking advantage of Rime's customization
+    options: spell, pauses, pronunciations, and inline speed control.
+
+### Deprecated
+
+- The TTS constructor field, `text_aggregator` is deprecated in favor of the new
+  `LLMTextProcessor`. TTSServices still have an internal aggregator for support
+  of default behavior, but if you want to override the aggregation behavior, you
+  should use the new processor.
+
+- The RTVI `bot-transcription` event is deprecated in favor of the new
+  `bot-output` message which is the canonical representation of bot output
+  (spoken or not). The code still emits a transcription message for backwards
+  compatibility while transition occurs.
+
+- Deprecated `add_pattern_pair` in the `PatternPairAggregator` which takes a
+  `pattern_id` and `remove_match` field in favor of the new `add_pattern` method
+  which takes a `type` and an `action`
+
+- `english_normalization` input parameter for `MiniMaxHttpTTSService` is
+  deprecated, use `test_normalization` instead.
+
 ### Fixed
 
+- Fixed an issue in `ElevenLabsRealtimeSTTService` where dynamic language
+  updates were not working.
+
+- Fixed an issue in `ElevenLabsRealtimeSTTService` where setting the sample
+  rate would result in transcripts failing.
+
 - Fixed `InworldTTSService` audio config payload to use camelCase keys expected
   by the Inworld API.
 
@@ -80,20 +314,11 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Updated language mappings for the Google and Gemini TTS services to match
   official documentation.
 
-- In `MiniMaxHttpTTSService`:
--- Added support for speech-2.6-hd and speech-2.6-turbo models
--- Added languages: Afrikaans, Bulgarian, Catalan, Danish, Persian, Filipino, Hebrew,
-Croatian, Hungarian, Malay, Norwegian, Nynorsk, Slovak, Slovenian, Swedish, and Tamil
--- Added new emotions: calm and fluent
-
 ### Deprecated
 
 - The `api_key` parameter in `GeminiTTSService` is deprecated. Use
   `credentials` or `credentials_path` instead for Google Cloud authentication.
 
-- `english_normalization` input parameter for `MiniMaxHttpTTSService` is deprecated,
-use `test_normalization` instead.
-
 ### Fixed
 
 - Fixed a `SimliVideoService` connection issue.

env.example

@@ -44,6 +44,7 @@ DAILY_SAMPLE_ROOM_URL=https://...
 
 # Deepgram
 DEEPGRAM_API_KEY=...
+SAGEMAKER_ENDPOINT_NAME=...
 
 # DeepSeek
 DEEPSEEK_API_KEY=...

examples/foundational/07c-interruptible-deepgram-sagemaker.py

@@ -0,0 +1,137 @@
+#
+# Copyright (c) 2024–2025, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+
+import os
+
+from dotenv import load_dotenv
+from loguru import logger
+
+from pipecat.audio.turn.smart_turn.base_smart_turn import SmartTurnParams
+from pipecat.audio.turn.smart_turn.local_smart_turn_v3 import LocalSmartTurnAnalyzerV3
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.audio.vad.vad_analyzer import VADParams
+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.aws.llm import AWSBedrockLLMService
+from pipecat.services.deepgram.stt_sagemaker import DeepgramSageMakerSTTService
+from pipecat.services.deepgram.tts import DeepgramTTSService
+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 store functions so objects (e.g. SileroVADAnalyzer) don't get
+# instantiated. The function will be called when the desired transport gets
+# selected.
+transport_params = {
+    "daily": lambda: DailyParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+        vad_analyzer=SileroVADAnalyzer(params=VADParams(stop_secs=0.2)),
+        turn_analyzer=LocalSmartTurnAnalyzerV3(params=SmartTurnParams()),
+    ),
+    "twilio": lambda: FastAPIWebsocketParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+        vad_analyzer=SileroVADAnalyzer(params=VADParams(stop_secs=0.2)),
+        turn_analyzer=LocalSmartTurnAnalyzerV3(params=SmartTurnParams()),
+    ),
+    "webrtc": lambda: TransportParams(
+        audio_in_enabled=True,
+        audio_out_enabled=True,
+        vad_analyzer=SileroVADAnalyzer(params=VADParams(stop_secs=0.2)),
+        turn_analyzer=LocalSmartTurnAnalyzerV3(params=SmartTurnParams()),
+    ),
+}
+
+
+async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
+    logger.info(f"Starting bot")
+
+    # Initialize Deepgram SageMaker STT Service
+    # This requires:
+    # - AWS credentials configured (via environment variables or AWS CLI)
+    # - A deployed SageMaker endpoint with Deepgram model
+    stt = DeepgramSageMakerSTTService(
+        endpoint_name=os.getenv("SAGEMAKER_ENDPOINT_NAME"),
+        region=os.getenv("AWS_REGION"),
+    )
+
+    tts = DeepgramTTSService(api_key=os.getenv("DEEPGRAM_API_KEY"), voice="aura-2-andromeda-en")
+
+    llm = AWSBedrockLLMService(
+        aws_region=os.getenv("AWS_REGION"),
+        model="us.amazon.nova-pro-v1:0",
+        params=AWSBedrockLLMService.InputParams(temperature=0.8),
+    )
+
+    messages = [
+        {
+            "role": "system",
+            "content": "You are a helpful LLM in a WebRTC call. Your goal is to demonstrate your capabilities in a succinct way. 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.",
+        },
+    ]
+
+    context = LLMContext(messages)
+    context_aggregator = LLMContextAggregatorPair(context)
+
+    pipeline = Pipeline(
+        [
+            transport.input(),  # Transport user input
+            stt,  # STT
+            context_aggregator.user(),  # User responses
+            llm,  # LLM
+            tts,  # TTS
+            transport.output(),  # Transport bot output
+            context_aggregator.assistant(),  # 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.
+        messages.append({"role": "system", "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/foundational/35-pattern-pair-voice-switching.py

@@ -62,7 +62,11 @@ 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.utils.text.pattern_pair_aggregator import PatternMatch, PatternPairAggregator
+from pipecat.utils.text.pattern_pair_aggregator import (
+    MatchAction,
+    PatternMatch,
+    PatternPairAggregator,
+)
 
 load_dotenv(override=True)
 
@@ -106,16 +110,16 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     pattern_aggregator = PatternPairAggregator()
 
     # Add pattern for voice switching
-    pattern_aggregator.add_pattern_pair(
+    pattern_aggregator.add_pattern(
-        pattern_id="voice_tag",
+        type="voice",
         start_pattern="<voice>",
         end_pattern="</voice>",
-        remove_match=True,
+        action=MatchAction.REMOVE,  # Remove tags from final text
     )
 
     # Register handler for voice switching
     async def on_voice_tag(match: PatternMatch):
-        voice_name = match.content.strip().lower()
+        voice_name = match.text.strip().lower()
         if voice_name in VOICE_IDS:
             # First flush any existing audio to finish the current context
             await tts.flush_audio()
@@ -125,7 +129,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
         else:
             logger.warning(f"Unknown voice: {voice_name}")
 
-    pattern_aggregator.on_pattern_match("voice_tag", on_voice_tag)
+    pattern_aggregator.on_pattern_match("voice", on_voice_tag)
 
     stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
 

pyproject.toml

@@ -50,13 +50,13 @@ anthropic = [ "anthropic~=0.49.0" ]
 assemblyai = [ "pipecat-ai[websockets-base]" ]
 asyncai = [ "pipecat-ai[websockets-base]" ]
 aws = [ "aioboto3~=15.0.0", "pipecat-ai[websockets-base]" ]
-aws-nova-sonic = [ "aws_sdk_bedrock_runtime~=0.1.1; python_version>='3.12'" ]
+aws-nova-sonic = [ "aws_sdk_bedrock_runtime~=0.2.0; python_version>='3.12'" ]
 azure = [ "azure-cognitiveservices-speech~=1.42.0"]
 cartesia = [ "cartesia~=2.0.3", "pipecat-ai[websockets-base]" ]
 cerebras = []
-deepseek = []
 daily = [ "daily-python~=0.22.0" ]
 deepgram = [ "deepgram-sdk~=4.7.0" ]
+deepseek = []
 elevenlabs = [ "pipecat-ai[websockets-base]" ]
 fal = [ "fal-client~=0.5.9" ]
 fireworks = []
@@ -69,19 +69,21 @@ gstreamer = [ "pygobject~=3.50.0" ]
 heygen = [ "livekit>=1.0.13", "pipecat-ai[websockets-base]" ]
 hume = [ "hume>=0.11.2" ]
 inworld = []
-krisp = [ "pipecat-ai-krisp~=0.4.0" ]
 koala = [ "pvkoala~=2.0.3" ]
+krisp = [ "pipecat-ai-krisp~=0.4.0" ]
 langchain = [ "langchain~=0.3.20", "langchain-community~=0.3.20", "langchain-openai~=0.3.9" ]
 livekit = [ "livekit~=1.0.13", "livekit-api~=1.0.5", "tenacity>=8.2.3,<10.0.0", "pyjwt>=2.10.1" ]
 lmnt = [ "pipecat-ai[websockets-base]" ]
 local = [ "pyaudio~=0.2.14" ]
+local-smart-turn = [ "coremltools>=8.0", "transformers", "torch>=2.5.0,<3", "torchaudio>=2.5.0,<3" ]
+local-smart-turn-v3 = [ "transformers", "onnxruntime>=1.20.1,<2" ]
 mcp = [ "mcp[cli]>=1.11.0,<2" ]
 mem0 = [ "mem0ai~=0.1.94" ]
 mistral = []
 mlx-whisper = [ "mlx-whisper~=0.4.2" ]
 moondream = [ "accelerate~=1.10.0", "einops~=0.8.0", "pyvips[binary]~=3.0.0", "timm~=1.0.13", "transformers>=4.48.0" ]
-nim = []
 neuphonic = [ "pipecat-ai[websockets-base]" ]
+nim = []
 noisereduce = [ "noisereduce~=3.0.3" ]
 openai = [ "pipecat-ai[websockets-base]" ]
 openpipe = [ "openpipe>=4.50.0,<6" ]
@@ -89,15 +91,14 @@ openrouter = []
 perplexity = []
 playht = [ "pipecat-ai[websockets-base]" ]
 qwen = []
+remote-smart-turn = []
 rime = [ "pipecat-ai[websockets-base]" ]
 riva = [ "nvidia-riva-client~=2.21.1" ]
 runner = [ "python-dotenv>=1.0.0,<2.0.0", "uvicorn>=0.32.0,<1.0.0", "fastapi>=0.115.6,<0.122.0", "pipecat-ai-small-webrtc-prebuilt>=1.0.0"]
+sagemaker = ["aws_sdk_sagemaker_runtime_http2; python_version>='3.12'"]
 sambanova = []
 sarvam = [ "sarvamai==0.1.21", "pipecat-ai[websockets-base]" ]
 sentry = [ "sentry-sdk>=2.28.0,<3" ]
-local-smart-turn = [ "coremltools>=8.0", "transformers", "torch>=2.5.0,<3", "torchaudio>=2.5.0,<3" ]
-local-smart-turn-v3 = [ "transformers", "onnxruntime>=1.20.1,<2" ]
-remote-smart-turn = []
 silero = [ "onnxruntime>=1.20.1,<2" ]
 simli = [ "simli-ai~=1.0.3"]
 soniox = [ "pipecat-ai[websockets-base]" ]

src/pipecat/extensions/ivr/ivr_navigator.py

@@ -31,7 +31,11 @@ from pipecat.pipeline.pipeline import Pipeline
 from pipecat.processors.aggregators.openai_llm_context import OpenAILLMContextFrame
 from pipecat.processors.frame_processor import FrameDirection, FrameProcessor
 from pipecat.services.llm_service import LLMService
-from pipecat.utils.text.pattern_pair_aggregator import PatternMatch, PatternPairAggregator
+from pipecat.utils.text.pattern_pair_aggregator import (
+    MatchAction,
+    PatternMatch,
+    PatternPairAggregator,
+)
 
 
 class IVRStatus(Enum):
@@ -114,15 +118,15 @@ class IVRProcessor(FrameProcessor):
     def _setup_xml_patterns(self):
         """Set up XML pattern detection and handlers."""
         # Register DTMF pattern
-        self._aggregator.add_pattern_pair("dtmf", "<dtmf>", "</dtmf>", remove_match=True)
+        self._aggregator.add_pattern("dtmf", "<dtmf>", "</dtmf>", action=MatchAction.REMOVE)
         self._aggregator.on_pattern_match("dtmf", self._handle_dtmf_action)
 
         # Register mode pattern
-        self._aggregator.add_pattern_pair("mode", "<mode>", "</mode>", remove_match=True)
+        self._aggregator.add_pattern("mode", "<mode>", "</mode>", action=MatchAction.REMOVE)
         self._aggregator.on_pattern_match("mode", self._handle_mode_action)
 
         # Register IVR pattern
-        self._aggregator.add_pattern_pair("ivr", "<ivr>", "</ivr>", remove_match=True)
+        self._aggregator.add_pattern("ivr", "<ivr>", "</ivr>", action=MatchAction.REMOVE)
         self._aggregator.on_pattern_match("ivr", self._handle_ivr_action)
 
     async def process_frame(self, frame: Frame, direction: FrameDirection):
@@ -148,7 +152,7 @@ class IVRProcessor(FrameProcessor):
             result = await self._aggregator.aggregate(frame.text)
             if result:
                 # Push aggregated text that doesn't contain XML patterns
-                await self.push_frame(LLMTextFrame(result), direction)
+                await self.push_frame(LLMTextFrame(result.text), direction)
 
         else:
             await self.push_frame(frame, direction)
@@ -159,7 +163,7 @@ class IVRProcessor(FrameProcessor):
         Args:
             match: The pattern match containing DTMF content.
         """
-        value = match.content
+        value = match.text
         logger.debug(f"DTMF detected: {value}")
 
         try:
@@ -180,7 +184,7 @@ class IVRProcessor(FrameProcessor):
         Args:
             match: The pattern match containing IVR status content.
         """
-        status = match.content
+        status = match.text
         logger.trace(f"IVR status detected: {status}")
 
         # Convert string to enum, with validation
@@ -211,7 +215,7 @@ class IVRProcessor(FrameProcessor):
         Args:
             match: The pattern match containing mode content.
         """
-        mode = match.content
+        mode = match.text
         logger.debug(f"Mode detected: {mode}")
         if mode == "conversation":
             await self._handle_conversation()

src/pipecat/frames/frames.py

@@ -12,6 +12,7 @@ and LLM processing.
 """
 
 from dataclasses import dataclass, field
+from enum import Enum
 from typing import (
     TYPE_CHECKING,
     Any,
@@ -337,11 +338,14 @@ class TextFrame(DataFrame):
     # mandatory fields of theirs to have defaults to preserve
     # non-default-before-default argument order)
     includes_inter_frame_spaces: bool = field(init=False)
+    # Whether this text frame should be appended to the LLM context.
+    append_to_context: bool = field(init=False)
 
     def __post_init__(self):
         super().__post_init__()
         self.skip_tts = False
         self.includes_inter_frame_spaces = False
+        self.append_to_context = True
 
     def __str__(self):
         pts = format_pts(self.pts)
@@ -358,8 +362,32 @@ class LLMTextFrame(TextFrame):
         self.includes_inter_frame_spaces = True
 
 
+class AggregationType(str, Enum):
+    """Built-in aggregation strings."""
+
+    SENTENCE = "sentence"
+    WORD = "word"
+
+    def __str__(self):
+        return self.value
+
+
 @dataclass
-class TTSTextFrame(TextFrame):
+class AggregatedTextFrame(TextFrame):
+    """Text frame representing an aggregation of TextFrames.
+
+    This frame contains multiple TextFrames aggregated together for processing
+    or output along with a field to indicate how they are aggregated.
+
+    Parameters:
+        aggregated_by: Method used to aggregate the text frames.
+    """
+
+    aggregated_by: AggregationType | str
+
+
+@dataclass
+class TTSTextFrame(AggregatedTextFrame):
     """Text frame generated by Text-to-Speech services."""
 
     pass

src/pipecat/processors/aggregators/llm_response.py

@@ -1001,7 +1001,7 @@ class LLMAssistantContextAggregator(LLMContextResponseAggregator):
         await self.push_aggregation()
 
     async def _handle_text(self, frame: TextFrame):
-        if not self._started:
+        if not self._started or not frame.append_to_context:
             return
 
         if self._params.expect_stripped_words:

src/pipecat/processors/aggregators/llm_response_universal.py

@@ -811,7 +811,7 @@ class LLMAssistantAggregator(LLMContextAggregator):
         await self.push_aggregation()
 
     async def _handle_text(self, frame: TextFrame):
-        if not self._started:
+        if not self._started or not frame.append_to_context:
             return
 
         # Make sure we really have text (spaces count, too!)

src/pipecat/processors/aggregators/llm_text_processor.py

@@ -0,0 +1,106 @@
+#
+# Copyright (c) 2024–2025, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""LLM text processor module for processing and aggregating raw LLM output text.
+
+This processor will convert LLMTextFrames into AggregatedTextFrames based on the
+configured text aggregator. Using the customizable aggregator, it provides
+functionality to handle or manipulate LLM text frames before they are sent to other
+components such as TTS services or context aggregators. It can be used to pre-aggregate
+and categorize, modify, or filter direct output tokens from the LLM.
+"""
+
+from typing import Optional
+
+from pipecat.frames.frames import (
+    AggregatedTextFrame,
+    EndFrame,
+    Frame,
+    InterruptionFrame,
+    LLMFullResponseEndFrame,
+    LLMTextFrame,
+)
+from pipecat.processors.frame_processor import FrameDirection, FrameProcessor
+from pipecat.utils.text.base_text_aggregator import BaseTextAggregator
+from pipecat.utils.text.simple_text_aggregator import SimpleTextAggregator
+
+
+class LLMTextProcessor(FrameProcessor):
+    """A processor for handling or manipulating LLM text frames before they are processed further.
+
+    This processor will convert LLMTextFrames into AggregatedTextFrames based on the configured
+    text aggregator. Using the customizable aggregator, it provides functionality to handle or
+    manipulate LLM text frames before they are sent to other components such as TTS services or
+    context aggregators. It can be used to pre-aggregate and categorize, modify, or filter direct
+    output tokens from the LLM.
+    """
+
+    def __init__(self, *, text_aggregator: Optional[BaseTextAggregator] = None, **kwargs):
+        """Initialize the LLM text processor.
+
+        Args:
+            text_aggregator: An optional text aggregator to use for processing LLM text frames. By
+                default, a SimpleTextAggregator aggregating by sentence will be used.
+            **kwargs: Additional arguments passed to parent class.
+
+        TODO: Allow transformations per aggregation type or all (and deprecate the TTS filters).
+        """
+        super().__init__(**kwargs)
+        self._text_aggregator: BaseTextAggregator = text_aggregator or SimpleTextAggregator()
+
+    async def process_frame(self, frame: Frame, direction: FrameDirection):
+        """Process an LLMTextFrames using the aggregator to generate AggregatedTextFrames.
+
+        Args:
+            frame: The frame to process.
+            direction: The direction of frame flow in the pipeline.
+        """
+        await super().process_frame(frame, direction)
+
+        if isinstance(frame, InterruptionFrame):
+            await self._handle_interruption(frame)
+            await self.push_frame(frame, direction)
+        elif isinstance(frame, LLMTextFrame):
+            await self._handle_llm_text(frame)
+        elif isinstance(frame, LLMFullResponseEndFrame):
+            await self._handle_llm_end(frame.skip_tts)
+            await self.push_frame(frame, direction)
+        elif isinstance(frame, EndFrame):
+            await self._handle_llm_end()
+            await self.push_frame(frame, direction)
+        else:
+            await self.push_frame(frame, direction)
+
+    async def _handle_interruption(self, _):
+        """Handle interruptions by resetting the text aggregator."""
+        await self._text_aggregator.handle_interruption()
+
+    async def reset(self):
+        """Reset the internal state of the text processor and its aggregator."""
+        await self._text_aggregator.reset()
+
+    async def _handle_llm_text(self, in_frame: LLMTextFrame):
+        aggregation = await self._text_aggregator.aggregate(in_frame.text)
+        if aggregation:
+            out_frame = AggregatedTextFrame(
+                text=aggregation.text,
+                aggregated_by=aggregation.type,
+            )
+            out_frame.skip_tts = in_frame.skip_tts
+            await self.push_frame(out_frame)
+
+    async def _handle_llm_end(self, skip_tts: bool = False):
+        # Flush any remaining aggregated text at the end of the LLM response
+        aggregation = self._text_aggregator.text
+        await self._text_aggregator.reset()
+        text = aggregation.text.strip()
+        if text:
+            out_frame = AggregatedTextFrame(
+                text=text,
+                aggregated_by=aggregation.type,
+            )
+            out_frame.skip_tts = skip_tts
+            await self.push_frame(out_frame)

src/pipecat/processors/frameworks/rtvi.py

@@ -24,6 +24,7 @@ from typing import (
     Literal,
     Mapping,
     Optional,
+    Tuple,
     Union,
 )
 
@@ -32,6 +33,8 @@ from pydantic import BaseModel, Field, PrivateAttr, ValidationError
 
 from pipecat.audio.utils import calculate_audio_volume
 from pipecat.frames.frames import (
+    AggregatedTextFrame,
+    AggregationType,
     BotStartedSpeakingFrame,
     BotStoppedSpeakingFrame,
     CancelFrame,
@@ -704,6 +707,29 @@ class RTVITextMessageData(BaseModel):
     text: str
 
 
+class RTVIBotOutputMessageData(RTVITextMessageData):
+    """Data for bot output RTVI messages.
+
+    Extends RTVITextMessageData to include metadata about the output.
+    """
+
+    spoken: bool = False  # Indicates if the text has been spoken by TTS
+    aggregated_by: AggregationType | str
+    # Indicates what form the text is in (e.g., by word, sentence, etc.)
+
+
+class RTVIBotOutputMessage(BaseModel):
+    """Message containing bot output text.
+
+    An event meant to holistically represent what the bot is outputting,
+    along with metadata about the output and if it has been spoken.
+    """
+
+    label: RTVIMessageLiteral = RTVI_MESSAGE_LABEL
+    type: Literal["bot-output"] = "bot-output"
+    data: RTVIBotOutputMessageData
+
+
 class RTVIBotTranscriptionMessage(BaseModel):
     """Message containing bot transcription text.
 
@@ -896,6 +922,7 @@ class RTVIObserverParams:
         Parameter `errors_enabled` is deprecated. Error messages are always enabled.
 
     Parameters:
+        bot_output_enabled: Indicates if bot output messages should be sent.
         bot_llm_enabled: Indicates if the bot's LLM messages should be sent.
         bot_tts_enabled: Indicates if the bot's TTS messages should be sent.
         bot_speaking_enabled: Indicates if the bot's started/stopped speaking messages should be sent.
@@ -907,9 +934,17 @@ class RTVIObserverParams:
         metrics_enabled: Indicates if metrics messages should be sent.
         system_logs_enabled: Indicates if system logs should be sent.
         errors_enabled: [Deprecated] Indicates if errors messages should be sent.
+        skip_aggregator_types: List of aggregation types to skip sending as tts/output messages.
+          Note: if using this to avoid sending secure information, be sure to also disable
+                bot_llm_enabled to avoid leaking through LLM messages.
+        bot_output_transforms: A list of callables to transform text before just before sending it
+            to TTS. Each callable takes the aggregated text and its type, and returns the
+            transformed text. To register, provide a list of tuples of
+            (aggregation_type | '*', transform_function).
         audio_level_period_secs: How often audio levels should be sent if enabled.
     """
 
+    bot_output_enabled: bool = True
     bot_llm_enabled: bool = True
     bot_tts_enabled: bool = True
     bot_speaking_enabled: bool = True
@@ -921,6 +956,15 @@ class RTVIObserverParams:
     metrics_enabled: bool = True
     system_logs_enabled: bool = False
     errors_enabled: Optional[bool] = None
+    skip_aggregator_types: Optional[List[AggregationType | str]] = None
+    bot_output_transforms: Optional[
+        List[
+            Tuple[
+                AggregationType | str,
+                Callable[[str, AggregationType | str], Awaitable[str]],
+            ]
+        ]
+    ] = None
     audio_level_period_secs: float = 0.15
 
 
@@ -973,8 +1017,45 @@ class RTVIObserver(BaseObserver):
                     DeprecationWarning,
                 )
 
+        self._aggregation_transforms: List[
+            Tuple[AggregationType | str, Callable[[str, AggregationType | str], Awaitable[str]]]
+        ] = self._params.bot_output_transforms or []
+
+    def add_bot_output_transformer(
+        self,
+        transform_function: Callable[[str, AggregationType | str], Awaitable[str]],
+        aggregation_type: AggregationType | str = "*",
+    ):
+        """Transform text for a specific aggregation type before sending as Bot Output or TTS.
+
+        Args:
+            transform_function: The function to apply for transformation. This function should take
+                the text and aggregation type as input and return the transformed text.
+                Ex.: async def my_transform(text: str, aggregation_type: str) -> str:
+            aggregation_type: The type of aggregation to transform. This value defaults to "*" to
+                handle all text before sending to the client.
+        """
+        self._aggregation_transforms.append((aggregation_type, transform_function))
+
+    def remove_bot_output_transformer(
+        self,
+        transform_function: Callable[[str, AggregationType | str], Awaitable[str]],
+        aggregation_type: AggregationType | str = "*",
+    ):
+        """Remove a text transformer for a specific aggregation type.
+
+        Args:
+            transform_function: The function to remove.
+            aggregation_type: The type of aggregation to remove the transformer for.
+        """
+        self._aggregation_transforms = [
+            (agg_type, func)
+            for agg_type, func in self._aggregation_transforms
+            if not (agg_type == aggregation_type and func == transform_function)
+        ]
+
     async def _logger_sink(self, message):
-        """Logger sink so we cna send system logs to RTVI clients."""
+        """Logger sink so we can send system logs to RTVI clients."""
         message = RTVISystemLogMessage(data=RTVITextMessageData(text=message))
         await self.send_rtvi_message(message)
 
@@ -1048,12 +1129,15 @@ class RTVIObserver(BaseObserver):
             await self.send_rtvi_message(RTVIBotTTSStartedMessage())
         elif isinstance(frame, TTSStoppedFrame) and self._params.bot_tts_enabled:
             await self.send_rtvi_message(RTVIBotTTSStoppedMessage())
-        elif isinstance(frame, TTSTextFrame) and self._params.bot_tts_enabled:
+        elif isinstance(frame, AggregatedTextFrame) and (
-            if isinstance(src, BaseOutputTransport):
+            self._params.bot_output_enabled or self._params.bot_tts_enabled
-                message = RTVIBotTTSTextMessage(data=RTVITextMessageData(text=frame.text))
+        ):
-                await self.send_rtvi_message(message)
+            if isinstance(frame, TTSTextFrame) and not isinstance(src, BaseOutputTransport):
-            else:
+                # This check is to make sure we handle the frame when it has gone
+                # through the transport and has correct timing.
                 mark_as_seen = False
+            else:
+                await self._handle_aggregated_llm_text(frame)
         elif isinstance(frame, MetricsFrame) and self._params.metrics_enabled:
             await self._handle_metrics(frame)
         elif isinstance(frame, RTVIServerMessageFrame):
@@ -1084,15 +1168,6 @@ class RTVIObserver(BaseObserver):
         if mark_as_seen:
             self._frames_seen.add(frame.id)
 
-    async def _push_bot_transcription(self):
-        """Push accumulated bot transcription as a message."""
-        if len(self._bot_transcription) > 0:
-            message = RTVIBotTranscriptionMessage(
-                data=RTVITextMessageData(text=self._bot_transcription)
-            )
-            await self.send_rtvi_message(message)
-            self._bot_transcription = ""
-
     async def _handle_interruptions(self, frame: Frame):
         """Handle user speaking interruption frames."""
         message = None
@@ -1115,14 +1190,45 @@ class RTVIObserver(BaseObserver):
         if message:
             await self.send_rtvi_message(message)
 
+    async def _handle_aggregated_llm_text(self, frame: AggregatedTextFrame):
+        """Handle aggregated LLM text output frames."""
+        # Skip certain aggregator types if configured to do so.
+        if (
+            self._params.skip_aggregator_types
+            and frame.aggregated_by in self._params.skip_aggregator_types
+        ):
+            return
+
+        text = frame.text
+        type = frame.aggregated_by
+        for aggregation_type, transform in self._aggregation_transforms:
+            if aggregation_type == type or aggregation_type == "*":
+                text = await transform(text, type)
+
+        isTTS = isinstance(frame, TTSTextFrame)
+        if self._params.bot_output_enabled:
+            message = RTVIBotOutputMessage(
+                data=RTVIBotOutputMessageData(text=text, spoken=isTTS, aggregated_by=type)
+            )
+            await self.send_rtvi_message(message)
+
+        if isTTS and self._params.bot_tts_enabled:
+            tts_message = RTVIBotTTSTextMessage(data=RTVITextMessageData(text=text))
+            await self.send_rtvi_message(tts_message)
+
     async def _handle_llm_text_frame(self, frame: LLMTextFrame):
         """Handle LLM text output frames."""
         message = RTVIBotLLMTextMessage(data=RTVITextMessageData(text=frame.text))
         await self.send_rtvi_message(message)
 
+        # TODO (mrkb): Remove all this logic when we fully deprecate bot-transcription messages.
         self._bot_transcription += frame.text
-        if match_endofsentence(self._bot_transcription):
+
-            await self._push_bot_transcription()
+        if match_endofsentence(self._bot_transcription) and len(self._bot_transcription) > 0:
+            await self.send_rtvi_message(
+                RTVIBotTranscriptionMessage(data=RTVITextMessageData(text=self._bot_transcription))
+            )
+            self._bot_transcription = ""
 
     async def _handle_user_transcriptions(self, frame: Frame):
         """Handle user transcription frames."""
@@ -1248,7 +1354,7 @@ class RTVIProcessor(FrameProcessor):
         # Default to 0.3.0 which is the last version before actually having a
         # "client-version".
         self._client_version = [0, 3, 0]
-        self._skip_tts: bool = False  # Keep in sync with llm_service.py
+        self._llm_skip_tts: bool = False  # Keep in sync with llm_service.py's configuration.
 
         self._registered_actions: Dict[str, RTVIAction] = {}
         self._registered_services: Dict[str, RTVIService] = {}
@@ -1441,7 +1547,7 @@ class RTVIProcessor(FrameProcessor):
         elif isinstance(frame, RTVIActionFrame):
             await self._action_queue.put(frame)
         elif isinstance(frame, LLMConfigureOutputFrame):
-            self._skip_tts = frame.skip_tts
+            self._llm_skip_tts = frame.skip_tts
             await self.push_frame(frame, direction)
         # Other frames
         else:
@@ -1697,9 +1803,9 @@ class RTVIProcessor(FrameProcessor):
         opts = data.options if data.options is not None else RTVISendTextOptions()
         if opts.run_immediately:
             await self.interrupt_bot()
-        cur_skip_tts = self._skip_tts
+        cur_llm_skip_tts = self._llm_skip_tts
         should_skip_tts = not opts.audio_response
-        toggle_skip_tts = cur_skip_tts != should_skip_tts
+        toggle_skip_tts = cur_llm_skip_tts != should_skip_tts
         if toggle_skip_tts:
             output_frame = LLMConfigureOutputFrame(skip_tts=should_skip_tts)
             await self.push_frame(output_frame)
@@ -1709,7 +1815,7 @@ class RTVIProcessor(FrameProcessor):
         )
         await self.push_frame(text_frame)
         if toggle_skip_tts:
-            output_frame = LLMConfigureOutputFrame(skip_tts=cur_skip_tts)
+            output_frame = LLMConfigureOutputFrame(skip_tts=cur_llm_skip_tts)
             await self.push_frame(output_frame)
 
     async def _handle_update_context(self, data: RTVIAppendToContextData):

src/pipecat/services/aws/__init__.py

@@ -10,6 +10,7 @@ from pipecat.services import DeprecatedModuleProxy
 
 from .llm import *
 from .nova_sonic import *
+from .sagemaker import *
 from .stt import *
 from .tts import *
 

src/pipecat/services/aws/nova_sonic/llm.py

@@ -27,6 +27,7 @@ from pydantic import BaseModel, Field
 from pipecat.adapters.schemas.tools_schema import ToolsSchema
 from pipecat.adapters.services.aws_nova_sonic_adapter import AWSNovaSonicLLMAdapter, Role
 from pipecat.frames.frames import (
+    AggregationType,
     BotStoppedSpeakingFrame,
     CancelFrame,
     EndFrame,
@@ -1027,7 +1028,7 @@ class AWSNovaSonicLLMService(LLMService):
         logger.debug(f"Assistant response text added: {text}")
 
         # Report the text of the assistant response.
-        frame = TTSTextFrame(text)
+        frame = TTSTextFrame(text, aggregated_by=AggregationType.SENTENCE)
         frame.includes_inter_frame_spaces = True
         await self.push_frame(frame)
 
@@ -1062,7 +1063,9 @@ class AWSNovaSonicLLMService(LLMService):
                 # TTSTextFrame would be ignored otherwise (the interruption frame
                 # would have cleared the assistant aggregator state).
                 await self.push_frame(LLMFullResponseStartFrame())
-                frame = TTSTextFrame(self._assistant_text_buffer)
+                frame = TTSTextFrame(
+                    self._assistant_text_buffer, aggregated_by=AggregationType.SENTENCE
+                )
                 frame.includes_inter_frame_spaces = True
                 await self.push_frame(frame)
             self._may_need_repush_assistant_text = False

src/pipecat/services/aws/sagemaker/bidi_client.py

@@ -0,0 +1,283 @@
+#
+# Copyright (c) 2024–2025, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""AWS SageMaker bidirectional streaming client.
+
+This module provides a client for streaming bidirectional communication with
+SageMaker endpoints using the HTTP/2 protocol. Supports sending audio, text,
+and JSON data to SageMaker model endpoints and receiving streaming responses.
+"""
+
+import os
+from typing import Optional
+
+from loguru import logger
+
+try:
+    from aws_sdk_sagemaker_runtime_http2.client import SageMakerRuntimeHTTP2Client
+    from aws_sdk_sagemaker_runtime_http2.config import Config, HTTPAuthSchemeResolver
+    from aws_sdk_sagemaker_runtime_http2.models import (
+        InvokeEndpointWithBidirectionalStreamInput,
+        RequestPayloadPart,
+        RequestStreamEventPayloadPart,
+        ResponseStreamEvent,
+    )
+    from smithy_aws_core.auth.sigv4 import SigV4AuthScheme
+    from smithy_aws_core.identity import EnvironmentCredentialsResolver
+    from smithy_core.aio.eventstream import DuplexEventStream
+except ModuleNotFoundError as e:
+    logger.error(f"Exception: {e}")
+    logger.error(
+        "In order to use SageMaker BiDi client, you need to `pip install pipecat-ai[sagemaker]`."
+    )
+    raise Exception(f"Missing module: {e}")
+
+
+class SageMakerBidiClient:
+    """Client for bidirectional streaming with AWS SageMaker endpoints.
+
+    Handles low-level HTTP/2 bidirectional streaming protocol for communicating
+    with SageMaker model endpoints. Provides methods for sending various data
+    types (audio, text, JSON) and receiving streaming responses.
+
+    This client uses AWS SigV4 authentication and supports credential resolution
+    from environment variables, AWS CLI configuration, and instance metadata.
+
+    Example::
+
+        client = SageMakerBidiClient(
+            endpoint_name="my-deepgram-endpoint",
+            region="us-east-2",
+            model_invocation_path="v1/listen",
+            model_query_string="model=nova-3&language=en"
+        )
+        await client.start_session()
+        await client.send_audio_chunk(audio_bytes)
+        response = await client.receive_response()
+        await client.close_session()
+    """
+
+    def __init__(
+        self,
+        endpoint_name: str,
+        region: str,
+        model_invocation_path: str = "",
+        model_query_string: str = "",
+    ):
+        """Initialize the SageMaker BiDi client.
+
+        Args:
+            endpoint_name: Name of the SageMaker endpoint to connect to.
+            region: AWS region where the endpoint is deployed.
+            model_invocation_path: API path for the model invocation (e.g., "v1/listen").
+            model_query_string: Query string parameters for the model (e.g., "model=nova-3").
+        """
+        self.endpoint_name = endpoint_name
+        self.region = region
+        self.model_invocation_path = model_invocation_path
+        self.model_query_string = model_query_string
+        self.bidi_endpoint = f"https://runtime.sagemaker.{region}.amazonaws.com:8443"
+        self._client: Optional[SageMakerRuntimeHTTP2Client] = None
+        self._stream: Optional[
+            DuplexEventStream[RequestStreamEventPayloadPart, ResponseStreamEvent, any]
+        ] = None
+        self._output_stream = None
+        self._is_active = False
+
+    def _initialize_client(self):
+        """Initialize the SageMaker Runtime HTTP2 client with AWS credentials.
+
+        Creates and configures the SageMaker Runtime HTTP2 client with SigV4
+        authentication. Attempts to resolve AWS credentials from environment
+        variables, AWS CLI configuration, or instance metadata.
+        """
+        logger.debug(f"Initializing SageMaker BiDi client for region: {self.region}")
+        logger.debug(f"Using endpoint URI: {self.bidi_endpoint}")
+
+        # Check for AWS credentials
+        has_env_creds = bool(os.getenv("AWS_ACCESS_KEY_ID") and os.getenv("AWS_SECRET_ACCESS_KEY"))
+
+        if not has_env_creds:
+            logger.warning(
+                "AWS credentials not found in environment variables. "
+                "Attempting to use EnvironmentCredentialsResolver which will check "
+                "AWS CLI configuration and instance metadata."
+            )
+
+        config = Config(
+            endpoint_uri=self.bidi_endpoint,
+            region=self.region,
+            aws_credentials_identity_resolver=EnvironmentCredentialsResolver(),
+            auth_scheme_resolver=HTTPAuthSchemeResolver(),
+            auth_schemes={"aws.auth#sigv4": SigV4AuthScheme(service="sagemaker")},
+        )
+        self._client = SageMakerRuntimeHTTP2Client(config=config)
+
+    async def start_session(self):
+        """Start a bidirectional streaming session with the SageMaker endpoint.
+
+        Initializes the client if needed, creates the bidirectional stream, and
+        establishes the connection to the SageMaker endpoint. Must be called
+        before sending or receiving data.
+
+        Returns:
+            The output stream for receiving responses.
+
+        Raises:
+            RuntimeError: If client initialization or connection fails.
+        """
+        if not self._client:
+            self._initialize_client()
+
+        logger.debug(f"Starting BiDi session with endpoint: {self.endpoint_name}")
+        logger.debug(f"Model invocation path: {self.model_invocation_path}")
+        logger.debug(f"Model query string: {self.model_query_string}")
+
+        # Create the bidirectional stream
+        stream_input = InvokeEndpointWithBidirectionalStreamInput(
+            endpoint_name=self.endpoint_name,
+            model_invocation_path=self.model_invocation_path,
+            model_query_string=self.model_query_string,
+        )
+
+        try:
+            self._stream = await self._client.invoke_endpoint_with_bidirectional_stream(
+                stream_input
+            )
+            self._is_active = True
+
+            # Get output stream
+            output = await self._stream.await_output()
+            self._output_stream = output[1]
+
+            logger.debug("BiDi session started successfully")
+            return self._output_stream
+
+        except Exception as e:
+            logger.error(f"Failed to start BiDi session: {e}")
+            self._is_active = False
+            raise RuntimeError(f"Failed to start SageMaker BiDi session: {e}")
+
+    async def send_data(self, data_bytes: bytes, data_type: Optional[str] = None):
+        """Send a chunk of data to the stream.
+
+        Generic method for sending any type of data to the SageMaker endpoint.
+        Use the convenience methods (send_audio_chunk, send_text, send_json)
+        for common data types.
+
+        Args:
+            data_bytes: Raw bytes to send.
+            data_type: Optional data type header. Common values are "BINARY" for
+                audio/binary data and "UTF8" for text/JSON data.
+
+        Raises:
+            RuntimeError: If session is not active or send fails.
+        """
+        if not self._is_active or not self._stream:
+            raise RuntimeError("BiDi session not active")
+
+        try:
+            payload = RequestPayloadPart(bytes_=data_bytes, data_type=data_type)
+            event = RequestStreamEventPayloadPart(value=payload)
+            await self._stream.input_stream.send(event)
+        except Exception as e:
+            logger.error(f"Failed to send data: {e}")
+            raise
+
+    async def send_audio_chunk(self, audio_bytes: bytes):
+        """Send a chunk of audio data to the stream.
+
+        Convenience method for sending audio data. Automatically sets the data
+        type to "BINARY".
+
+        Args:
+            audio_bytes: Raw audio bytes to send (e.g., PCM audio data).
+
+        Raises:
+            RuntimeError: If session is not active or send fails.
+        """
+        await self.send_data(audio_bytes, data_type="BINARY")
+
+    async def send_text(self, text: str):
+        """Send text data to the stream.
+
+        Convenience method for sending text data. Automatically encodes the text
+        as UTF-8 and sets the data type to "UTF8".
+
+        Args:
+            text: Text string to send.
+
+        Raises:
+            RuntimeError: If session is not active or send fails.
+        """
+        await self.send_data(text.encode("utf-8"), data_type="UTF8")
+
+    async def send_json(self, data: dict):
+        """Send JSON data to the stream.
+
+        Convenience method for sending JSON-encoded messages. Useful for control
+        messages like KeepAlive or CloseStream. Automatically serializes the
+        dictionary to JSON, encodes as UTF-8, and sets the data type to "UTF8".
+
+        Args:
+            data: Dictionary to send as JSON (e.g., {"type": "KeepAlive"}).
+
+        Raises:
+            RuntimeError: If session is not active or send fails.
+        """
+        import json
+
+        await self.send_data(json.dumps(data).encode("utf-8"), data_type="UTF8")
+
+    async def receive_response(self) -> Optional[ResponseStreamEvent]:
+        """Receive a response from the stream.
+
+        Blocks until a response is available from the SageMaker endpoint. Returns
+        None when the stream is closed.
+
+        Returns:
+            The response event containing payload data, or None if stream is closed.
+
+        Raises:
+            RuntimeError: If session is not active.
+        """
+        if not self._is_active or not self._output_stream:
+            raise RuntimeError("BiDi session not active")
+
+        try:
+            result = await self._output_stream.receive()
+            return result
+        except Exception as e:
+            logger.error(f"Failed to receive response: {e}")
+            raise
+
+    async def close_session(self):
+        """Close the bidirectional streaming session.
+
+        Gracefully closes the input stream and marks the session as inactive.
+        Safe to call multiple times.
+        """
+        if not self._is_active:
+            return
+
+        logger.debug("Closing BiDi session...")
+        self._is_active = False
+
+        try:
+            if self._stream:
+                await self._stream.input_stream.close()
+            logger.debug("BiDi session closed successfully")
+        except Exception as e:
+            logger.warning(f"Error closing BiDi session: {e}")
+
+    @property
+    def is_active(self) -> bool:
+        """Check if the session is currently active.
+
+        Returns:
+            True if session is active, False otherwise.
+        """
+        return self._is_active

src/pipecat/services/cartesia/tts.py

@@ -10,7 +10,8 @@ import base64
 import json
 import uuid
 import warnings
-from typing import AsyncGenerator, List, Literal, Optional, Union
+from enum import Enum
+from typing import AsyncGenerator, List, Literal, Optional
 
 from loguru import logger
 from pydantic import BaseModel, Field
@@ -125,6 +126,72 @@ def language_to_cartesia_language(language: Language) -> Optional[str]:
     return resolve_language(language, LANGUAGE_MAP, use_base_code=True)
 
 
+class CartesiaEmotion(str, Enum):
+    """Predefined Emotions supported by Cartesia."""
+
+    # Primary emotions supported by Cartesia
+    NEUTRAL = "neutral"
+    ANGRY = "angry"
+    EXCITED = "excited"
+    CONTENT = "content"
+    SAD = "sad"
+    SCARED = "scared"
+    # Additional emotions supported by Cartesia
+    HAPPY = "happy"
+    ENTHUSIASTIC = "enthusiastic"
+    ELATED = "elated"
+    EUPHORIC = "euphoric"
+    TRIUMPHANT = "triumphant"
+    AMAZED = "amazed"
+    SURPRISED = "surprised"
+    FLIRTATIOUS = "flirtatious"
+    JOKING_COMEDIC = "joking/comedic"
+    CURIOUS = "curious"
+    PEACEFUL = "peaceful"
+    SERENE = "serene"
+    CALM = "calm"
+    GRATEFUL = "grateful"
+    AFFECTIONATE = "affectionate"
+    TRUST = "trust"
+    SYMPATHETIC = "sympathetic"
+    ANTICIPATION = "anticipation"
+    MYSTERIOUS = "mysterious"
+    MAD = "mad"
+    OUTRAGED = "outraged"
+    FRUSTRATED = "frustrated"
+    AGITATED = "agitated"
+    THREATENED = "threatened"
+    DISGUSTED = "disgusted"
+    CONTEMPT = "contempt"
+    ENVIOUS = "envious"
+    SARCASTIC = "sarcastic"
+    IRONIC = "ironic"
+    DEJECTED = "dejected"
+    MELANCHOLIC = "melancholic"
+    DISAPPOINTED = "disappointed"
+    HURT = "hurt"
+    GUILTY = "guilty"
+    BORED = "bored"
+    TIRED = "tired"
+    REJECTED = "rejected"
+    NOSTALGIC = "nostalgic"
+    WISTFUL = "wistful"
+    APOLOGETIC = "apologetic"
+    HESITANT = "hesitant"
+    INSECURE = "insecure"
+    CONFUSED = "confused"
+    RESIGNED = "resigned"
+    ANXIOUS = "anxious"
+    PANICKED = "panicked"
+    ALARMED = "alarmed"
+    PROUD = "proud"
+    CONFIDENT = "confident"
+    DISTANT = "distant"
+    SKEPTICAL = "skeptical"
+    CONTEMPLATIVE = "contemplative"
+    DETERMINED = "determined"
+
+
 class CartesiaTTSService(AudioContextWordTTSService):
     """Cartesia TTS service with WebSocket streaming and word timestamps.
 
@@ -182,6 +249,10 @@ class CartesiaTTSService(AudioContextWordTTSService):
             container: Audio container format.
             params: Additional input parameters for voice customization.
             text_aggregator: Custom text aggregator for processing input text.
+
+                .. deprecated:: 0.0.95
+                    Use an LLMTextProcessor before the TTSService for custom text aggregation.
+
             aggregate_sentences: Whether to aggregate sentences within the TTSService.
             **kwargs: Additional arguments passed to the parent service.
         """
@@ -200,10 +271,18 @@ class CartesiaTTSService(AudioContextWordTTSService):
             push_text_frames=False,
             pause_frame_processing=True,
             sample_rate=sample_rate,
-            text_aggregator=text_aggregator or SkipTagsAggregator([("<spell>", "</spell>")]),
+            text_aggregator=text_aggregator,
             **kwargs,
         )
 
+        if not text_aggregator:
+            # Always skip tags added for spelled-out text
+            # Note: This is primarily to support backwards compatibility.
+            #    The preferred way of taking advantage of Cartesia SSML Tags is
+            #    to use an LLMTextProcessor and/or a text_transformer to identify
+            #    and insert these tags for the purpose of the TTS service alone.
+            self._text_aggregator = SkipTagsAggregator([("<spell>", "</spell>")])
+
         params = params or CartesiaTTSService.InputParams()
 
         self._api_key = api_key
@@ -257,6 +336,27 @@ class CartesiaTTSService(AudioContextWordTTSService):
         """
         return language_to_cartesia_language(language)
 
+    # A set of Cartesia-specific helpers for text transformations
+    def SPELL(text: str) -> str:
+        """Wrap text in Cartesia spell tag."""
+        return f"<spell>{text}</spell>"
+
+    def EMOTION_TAG(emotion: CartesiaEmotion) -> str:
+        """Convenience method to create an emotion tag."""
+        return f'<emotion value="{emotion}" />'
+
+    def PAUSE_TAG(seconds: float) -> str:
+        """Convenience method to create a pause tag."""
+        return f'<break time="{seconds}s" />'
+
+    def VOLUME_TAG(volume: float) -> str:
+        """Convenience method to create a volume tag."""
+        return f'<volume ratio="{volume}" />'
+
+    def SPEED_TAG(speed: float) -> str:
+        """Convenience method to create a speed tag."""
+        return f'<speed ratio="{speed}" />'
+
     def _is_cjk_language(self, language: str) -> bool:
         """Check if the given language is CJK (Chinese, Japanese, Korean).
 

src/pipecat/services/deepgram/__init__.py

@@ -10,6 +10,7 @@ from pipecat.services import DeprecatedModuleProxy
 
 from .flux import *
 from .stt import *
+from .stt_sagemaker import *
 from .tts import *
 
 sys.modules[__name__] = DeprecatedModuleProxy(globals(), "deepgram", "deepgram.[stt,tts]")

src/pipecat/services/deepgram/stt_sagemaker.py

@@ -0,0 +1,447 @@
+#
+# Copyright (c) 2024–2025, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Deepgram speech-to-text service for AWS SageMaker.
+
+This module provides a Pipecat STT service that connects to Deepgram models
+deployed on AWS SageMaker endpoints. Uses HTTP/2 bidirectional streaming for
+low-latency real-time transcription with support for interim results, multiple
+languages, and various Deepgram features.
+"""
+
+import asyncio
+import json
+from typing import AsyncGenerator, Optional
+
+from loguru import logger
+
+from pipecat.frames.frames import (
+    CancelFrame,
+    EndFrame,
+    ErrorFrame,
+    Frame,
+    InterimTranscriptionFrame,
+    StartFrame,
+    TranscriptionFrame,
+    UserStartedSpeakingFrame,
+    UserStoppedSpeakingFrame,
+)
+from pipecat.processors.frame_processor import FrameDirection
+from pipecat.services.aws.sagemaker.bidi_client import SageMakerBidiClient
+from pipecat.services.stt_service import STTService
+from pipecat.transcriptions.language import Language
+from pipecat.utils.time import time_now_iso8601
+from pipecat.utils.tracing.service_decorators import traced_stt
+
+try:
+    from deepgram import LiveOptions
+except ModuleNotFoundError as e:
+    logger.error(f"Exception: {e}")
+    logger.error(
+        "In order to use DeepgramSageMakerSTTService, you need to `pip install pipecat-ai[deepgram,sagemaker]`."
+    )
+    raise Exception(f"Missing module: {e}")
+
+
+class DeepgramSageMakerSTTService(STTService):
+    """Deepgram speech-to-text service for AWS SageMaker.
+
+    Provides real-time speech recognition using Deepgram models deployed on
+    AWS SageMaker endpoints. Uses HTTP/2 bidirectional streaming for low-latency
+    transcription with support for interim results, speaker diarization, and
+    multiple languages.
+
+    Requirements:
+
+    - AWS credentials configured (via environment variables, AWS CLI, or instance metadata)
+    - A deployed SageMaker endpoint with Deepgram model: https://developers.deepgram.com/docs/deploy-amazon-sagemaker
+    - Deepgram SDK for LiveOptions configuration
+
+    Example::
+
+        stt = DeepgramSageMakerSTTService(
+            endpoint_name="my-deepgram-endpoint",
+            region="us-east-2",
+            live_options=LiveOptions(
+                model="nova-3",
+                language="en",
+                interim_results=True,
+                punctuate=True,
+            ),
+        )
+    """
+
+    def __init__(
+        self,
+        *,
+        endpoint_name: str,
+        region: str,
+        sample_rate: Optional[int] = None,
+        live_options: Optional[LiveOptions] = None,
+        **kwargs,
+    ):
+        """Initialize the Deepgram SageMaker STT service.
+
+        Args:
+            endpoint_name: Name of the SageMaker endpoint with Deepgram model
+                deployed (e.g., "my-deepgram-nova-3-endpoint").
+            region: AWS region where the endpoint is deployed (e.g., "us-east-2").
+            sample_rate: Audio sample rate in Hz. If None, uses value from
+                live_options or defaults to the value from StartFrame.
+            live_options: Deepgram LiveOptions for detailed configuration. If None,
+                uses sensible defaults (nova-3 model, English, interim results enabled).
+            **kwargs: Additional arguments passed to the parent STTService.
+        """
+        sample_rate = sample_rate or (live_options.sample_rate if live_options else None)
+        super().__init__(sample_rate=sample_rate, **kwargs)
+
+        self._endpoint_name = endpoint_name
+        self._region = region
+
+        # Create default options similar to DeepgramSTTService
+        default_options = LiveOptions(
+            encoding="linear16",
+            language=Language.EN,
+            model="nova-3",
+            channels=1,
+            interim_results=True,
+            punctuate=True,
+        )
+
+        # Merge with provided options
+        merged_options = default_options.to_dict()
+        if live_options:
+            default_model = default_options.model
+            merged_options.update(live_options.to_dict())
+            # Handle the "None" string bug from deepgram-sdk
+            if "model" in merged_options and merged_options["model"] == "None":
+                merged_options["model"] = default_model
+
+        # Convert Language enum to string if needed
+        if "language" in merged_options and isinstance(merged_options["language"], Language):
+            merged_options["language"] = merged_options["language"].value
+
+        self.set_model_name(merged_options["model"])
+        self._settings = merged_options
+
+        self._client: Optional[SageMakerBidiClient] = None
+        self._response_task: Optional[asyncio.Task] = None
+        self._keepalive_task: Optional[asyncio.Task] = None
+
+    def can_generate_metrics(self) -> bool:
+        """Check if this service can generate processing metrics.
+
+        Returns:
+            True, as Deepgram SageMaker service supports metrics generation.
+        """
+        return True
+
+    async def set_model(self, model: str):
+        """Set the Deepgram model and reconnect.
+
+        Disconnects from the current session, updates the model setting, and
+        establishes a new connection with the updated model.
+
+        Args:
+            model: The Deepgram model name to use (e.g., "nova-3").
+        """
+        await super().set_model(model)
+        logger.info(f"Switching STT model to: [{model}]")
+        self._settings["model"] = model
+        await self._disconnect()
+        await self._connect()
+
+    async def set_language(self, language: Language):
+        """Set the recognition language and reconnect.
+
+        Disconnects from the current session, updates the language setting, and
+        establishes a new connection with the updated language.
+
+        Args:
+            language: The language to use for speech recognition (e.g., Language.EN,
+                Language.ES).
+        """
+        logger.info(f"Switching STT language to: [{language}]")
+        self._settings["language"] = language
+        await self._disconnect()
+        await self._connect()
+
+    async def start(self, frame: StartFrame):
+        """Start the Deepgram SageMaker STT service.
+
+        Args:
+            frame: The start frame containing initialization parameters.
+        """
+        await super().start(frame)
+        self._settings["sample_rate"] = self.sample_rate
+        await self._connect()
+
+    async def stop(self, frame: EndFrame):
+        """Stop the Deepgram SageMaker STT service.
+
+        Args:
+            frame: The end frame.
+        """
+        await super().stop(frame)
+        await self._disconnect()
+
+    async def cancel(self, frame: CancelFrame):
+        """Cancel the Deepgram SageMaker STT service.
+
+        Args:
+            frame: The cancel frame.
+        """
+        await super().cancel(frame)
+        await self._disconnect()
+
+    async def run_stt(self, audio: bytes) -> AsyncGenerator[Frame, None]:
+        """Send audio data to Deepgram for transcription.
+
+        Args:
+            audio: Raw audio bytes to transcribe.
+
+        Yields:
+            Frame: None (transcription results come via BiDi stream callbacks).
+        """
+        if self._client and self._client.is_active:
+            try:
+                await self._client.send_audio_chunk(audio)
+            except Exception as e:
+                logger.error(f"Error sending audio to SageMaker: {e}")
+                await self.push_error(ErrorFrame(error=f"SageMaker STT error: {e}"))
+        yield None
+
+    async def _connect(self):
+        """Connect to the SageMaker endpoint and start the BiDi session.
+
+        Builds the Deepgram query string from settings, creates the BiDi client,
+        starts the streaming session, and launches background tasks for processing
+        responses and sending KeepAlive messages.
+        """
+        logger.debug("Connecting to Deepgram on SageMaker...")
+
+        # Update sample rate in settings
+        self._settings["sample_rate"] = self.sample_rate
+
+        # Build query string from settings, converting booleans to strings
+        query_params = {}
+        for key, value in self._settings.items():
+            if value is not None:
+                # Convert boolean values to lowercase strings for Deepgram API
+                if isinstance(value, bool):
+                    query_params[key] = str(value).lower()
+                else:
+                    query_params[key] = str(value)
+
+        query_string = "&".join(f"{k}={v}" for k, v in query_params.items())
+
+        # Create BiDi client
+        self._client = SageMakerBidiClient(
+            endpoint_name=self._endpoint_name,
+            region=self._region,
+            model_invocation_path="v1/listen",
+            model_query_string=query_string,
+        )
+
+        try:
+            # Start the session
+            await self._client.start_session()
+
+            # Start processing responses in the background
+            self._response_task = self.create_task(self._process_responses())
+
+            # Start keepalive task to maintain connection
+            self._keepalive_task = self.create_task(self._send_keepalive())
+
+            logger.debug("Connected to Deepgram on SageMaker")
+            await self._call_event_handler("on_connected")
+
+        except Exception as e:
+            logger.error(f"Failed to connect to SageMaker: {e}")
+            await self.push_error(ErrorFrame(error=f"SageMaker connection error: {e}"))
+            await self._call_event_handler("on_connection_error", str(e))
+
+    async def _disconnect(self):
+        """Disconnect from the SageMaker endpoint.
+
+        Sends a CloseStream message to Deepgram, cancels background tasks
+        (KeepAlive and response processing), and closes the BiDi session.
+        Safe to call multiple times.
+        """
+        if self._client and self._client.is_active:
+            logger.debug("Disconnecting from Deepgram on SageMaker...")
+
+            # Send CloseStream message to Deepgram
+            try:
+                await self._client.send_json({"type": "CloseStream"})
+            except Exception as e:
+                logger.warning(f"Failed to send CloseStream message: {e}")
+
+            # Cancel keepalive task
+            if self._keepalive_task and not self._keepalive_task.done():
+                await self.cancel_task(self._keepalive_task)
+
+            # Cancel response processing task
+            if self._response_task and not self._response_task.done():
+                await self.cancel_task(self._response_task)
+
+            # Close the BiDi session
+            await self._client.close_session()
+
+            logger.debug("Disconnected from Deepgram on SageMaker")
+            await self._call_event_handler("on_disconnected")
+
+    async def _send_keepalive(self):
+        """Send periodic KeepAlive messages to maintain the connection.
+
+        Sends a KeepAlive JSON message to Deepgram every 5 seconds while the
+        connection is active. This prevents the connection from timing out during
+        periods of silence.
+        """
+        while self._client and self._client.is_active:
+            await asyncio.sleep(5)
+            if self._client and self._client.is_active:
+                try:
+                    await self._client.send_json({"type": "KeepAlive"})
+                except Exception as e:
+                    logger.warning(f"Failed to send KeepAlive: {e}")
+
+    async def _process_responses(self):
+        """Process streaming responses from Deepgram on SageMaker.
+
+        Continuously receives responses from the BiDi stream, decodes the payload,
+        parses JSON responses from Deepgram, and processes transcription results.
+        Runs as a background task until the connection is closed or cancelled.
+        """
+        try:
+            while self._client and self._client.is_active:
+                result = await self._client.receive_response()
+
+                if result is None:
+                    break
+
+                # Check if this is a PayloadPart with bytes
+                if hasattr(result, "value") and hasattr(result.value, "bytes_"):
+                    if result.value.bytes_:
+                        response_data = result.value.bytes_.decode("utf-8")
+
+                        try:
+                            # Parse JSON response from Deepgram
+                            parsed = json.loads(response_data)
+
+                            # Extract and process transcript if available
+                            if "channel" in parsed:
+                                await self._handle_transcript_response(parsed)
+
+                        except json.JSONDecodeError:
+                            logger.warning(f"Non-JSON response: {response_data}")
+
+        except asyncio.CancelledError:
+            logger.debug("Response processor cancelled")
+        except Exception as e:
+            logger.error(f"Error processing responses: {e}", exc_info=True)
+            await self.push_error(ErrorFrame(error=f"SageMaker response error: {e}"))
+        finally:
+            logger.debug("Response processor stopped")
+
+    async def _handle_transcript_response(self, parsed: dict):
+        """Handle a transcript response from Deepgram.
+
+        Extracts the transcript text, determines if it's final or interim, extracts
+        language information, and pushes the appropriate frame (TranscriptionFrame
+        or InterimTranscriptionFrame) downstream.
+
+        Args:
+            parsed: The parsed JSON response from Deepgram containing channel,
+                alternatives, transcript, and metadata.
+        """
+        alternatives = parsed.get("channel", {}).get("alternatives", [])
+        if not alternatives or not alternatives[0].get("transcript"):
+            return
+
+        transcript = alternatives[0]["transcript"]
+        if not transcript.strip():
+            return
+
+        # Stop TTFB metrics on first transcript
+        await self.stop_ttfb_metrics()
+
+        is_final = parsed.get("is_final", False)
+        speech_final = parsed.get("speech_final", False)
+
+        # Extract language if available
+        language = None
+        if alternatives[0].get("languages"):
+            language = alternatives[0]["languages"][0]
+            language = Language(language)
+
+        if is_final and speech_final:
+            # Final transcription
+            await self.push_frame(
+                TranscriptionFrame(
+                    transcript,
+                    self._user_id,
+                    time_now_iso8601(),
+                    language,
+                    result=parsed,
+                )
+            )
+            await self._handle_transcription(transcript, is_final, language)
+            await self.stop_processing_metrics()
+        else:
+            # Interim transcription
+            await self.push_frame(
+                InterimTranscriptionFrame(
+                    transcript,
+                    self._user_id,
+                    time_now_iso8601(),
+                    language,
+                    result=parsed,
+                )
+            )
+
+    @traced_stt
+    async def _handle_transcription(
+        self, transcript: str, is_final: bool, language: Optional[Language] = None
+    ):
+        """Handle a transcription result with tracing.
+
+        This method is decorated with @traced_stt for observability and tracing
+        integration. The actual transcription processing is handled by the parent
+        class and observers.
+
+        Args:
+            transcript: The transcribed text.
+            is_final: Whether this is a final transcription result.
+            language: The detected language of the transcription, if available.
+        """
+        pass
+
+    async def start_metrics(self):
+        """Start TTFB and processing metrics collection."""
+        await self.start_ttfb_metrics()
+        await self.start_processing_metrics()
+
+    async def process_frame(self, frame: Frame, direction: FrameDirection):
+        """Process frames with Deepgram SageMaker-specific handling.
+
+        Args:
+            frame: The frame to process.
+            direction: The direction of frame processing.
+        """
+        await super().process_frame(frame, direction)
+
+        # Start metrics when user starts speaking (if VAD is not provided by Deepgram)
+        if isinstance(frame, UserStartedSpeakingFrame):
+            await self.start_metrics()
+        elif isinstance(frame, UserStoppedSpeakingFrame):
+            # Send finalize message to Deepgram when user stops speaking
+            # This tells Deepgram to flush any remaining audio and return final results
+            if self._client and self._client.is_active:
+                try:
+                    await self._client.send_json({"type": "Finalize"})
+                except Exception as e:
+                    logger.warning(f"Error sending Finalize message: {e}")

src/pipecat/services/elevenlabs/stt.py

@@ -416,6 +416,8 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
                 Only used when commit_strategy is VAD. None uses ElevenLabs default.
             min_silence_duration_ms: Minimum silence duration for VAD (50-2000ms).
                 Only used when commit_strategy is VAD. None uses ElevenLabs default.
+            include_timestamps: Whether to include word-level timestamps in transcripts.
+            enable_logging: Whether to enable logging on ElevenLabs' side.
         """
 
         language_code: Optional[str] = None
@@ -424,6 +426,8 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
         vad_threshold: Optional[float] = None
         min_speech_duration_ms: Optional[int] = None
         min_silence_duration_ms: Optional[int] = None
+        include_timestamps: bool = False
+        enable_logging: bool = False
 
     def __init__(
         self,
@@ -459,6 +463,8 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
         self._audio_format = ""  # initialized in start()
         self._receive_task = None
 
+        self._settings = {"language": params.language_code}
+
     def can_generate_metrics(self) -> bool:
         """Check if the service can generate processing metrics.
 
@@ -477,7 +483,13 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
             Changing language requires reconnecting to the WebSocket.
         """
         logger.info(f"Switching STT language to: [{language}]")
-        self._params.language_code = language.value if isinstance(language, Language) else language
+        new_language = (
+            language_to_elevenlabs_language(language)
+            if isinstance(language, Language)
+            else language
+        )
+        self._params.language_code = new_language
+        self._settings["language"] = new_language
         # Reconnect with new settings
         await self._disconnect()
         await self._connect()
@@ -620,10 +632,16 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
             if self._params.language_code:
                 params.append(f"language_code={self._params.language_code}")
 
-            params.append(f"encoding={self._audio_format}")
+            params.append(f"audio_format={self._audio_format}")
-            params.append(f"sample_rate={self.sample_rate}")
             params.append(f"commit_strategy={self._params.commit_strategy.value}")
 
+            # Add optional parameters
+            if self._params.include_timestamps:
+                params.append(f"include_timestamps={str(self._params.include_timestamps).lower()}")
+
+            if self._params.enable_logging:
+                params.append(f"enable_logging={str(self._params.enable_logging).lower()}")
+
             # Add VAD parameters if using VAD commit strategy and values are specified
             if self._params.commit_strategy == CommitStrategy.VAD:
                 if self._params.vad_silence_threshold_secs is not None:
@@ -712,15 +730,20 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
         elif message_type == "committed_transcript_with_timestamps":
             await self._on_committed_transcript_with_timestamps(data)
 
-        elif message_type == "input_error":
+        elif message_type == "error":
-            error_msg = data.get("error", "Unknown input error")
+            error_msg = data.get("error", "Unknown error")
-            logger.error(f"ElevenLabs input error: {error_msg}")
+            logger.error(f"ElevenLabs error: {error_msg}")
-            await self.push_error(ErrorFrame(f"Input error: {error_msg}"))
+            await self.push_error(ErrorFrame(f"Error: {error_msg}"))
 
-        elif message_type in ["auth_error", "quota_exceeded", "transcriber_error", "error"]:
+        elif message_type == "auth_error":
-            error_msg = data.get("error", data.get("message", "Unknown error"))
+            error_msg = data.get("error", "Authentication error")
-            logger.error(f"ElevenLabs error ({message_type}): {error_msg}")
+            logger.error(f"ElevenLabs auth error: {error_msg}")
-            await self.push_error(ErrorFrame(f"{message_type}: {error_msg}"))
+            await self.push_error(ErrorFrame(f"Auth error: {error_msg}"))
+
+        elif message_type == "quota_exceeded_error":
+            error_msg = data.get("error", "Quota exceeded")
+            logger.error(f"ElevenLabs quota exceeded: {error_msg}")
+            await self.push_error(ErrorFrame(f"Quota exceeded: {error_msg}"))
 
         else:
             logger.debug(f"Unknown message type: {message_type}")
@@ -765,6 +788,11 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
         Args:
             data: Committed transcript data.
         """
+        # If timestamps are enabled, skip this message and wait for the
+        # committed_transcript_with_timestamps message which contains all the data
+        if self._params.include_timestamps:
+            return
+
         text = data.get("text", "").strip()
         if not text:
             return
@@ -792,6 +820,18 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
     async def _on_committed_transcript_with_timestamps(self, data: dict):
         """Handle committed transcript with word-level timestamps.
 
+        This message is sent when include_timestamps=true. The result data includes:
+        - text: The transcribed text
+        - language_code: Detected language (if available)
+        - words: Array of word objects with timing information:
+            - text: The word text
+            - start: Start time in seconds
+            - end: End time in seconds
+            - type: "word" or "spacing"
+            - speaker_id: Speaker identifier (if available)
+            - logprob: Log probability score (if available)
+            - characters: Array of character strings (if available)
+
         Args:
             data: Committed transcript data with timestamps.
         """
@@ -799,9 +839,24 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
         if not text:
             return
 
-        logger.debug(f"Committed transcript with timestamps: [{text}]")
+        await self.stop_ttfb_metrics()
-        logger.trace(f"Timestamps: {data.get('words', [])}")
+        await self.stop_processing_metrics()
 
-        # This is sent after the committed_transcript, so we don't need to
+        # Get language if provided
-        # push another TranscriptionFrame, but we could use the timestamps
+        language = data.get("language_code")
-        # for additional processing if needed in the future
+
+        logger.debug(f"Committed transcript with timestamps: [{text}]")
+
+        await self._handle_transcription(text, True, language)
+
+        # This message is sent after committed_transcript when include_timestamps=true.
+        # It contains the full transcript data including text and word-level timestamps.
+        await self.push_frame(
+            TranscriptionFrame(
+                text,
+                self._user_id,
+                time_now_iso8601(),
+                language,
+                result=data,
+            )
+        )

src/pipecat/services/google/gemini_live/llm.py

@@ -27,6 +27,7 @@ from pydantic import BaseModel, Field
 from pipecat.adapters.schemas.tools_schema import ToolsSchema
 from pipecat.adapters.services.gemini_adapter import GeminiLLMAdapter
 from pipecat.frames.frames import (
+    AggregationType,
     BotStartedSpeakingFrame,
     BotStoppedSpeakingFrame,
     CancelFrame,
@@ -1644,7 +1645,7 @@ class GeminiLiveLLMService(LLMService):
             await self.push_frame(TTSStartedFrame())
             await self.push_frame(LLMFullResponseStartFrame())
 
-        frame = TTSTextFrame(text=text)
+        frame = TTSTextFrame(text=text, aggregated_by=AggregationType.SENTENCE)
         # Gemini Live text already includes any necessary inter-chunk spaces
         frame.includes_inter_frame_spaces = True
 

src/pipecat/services/openai/realtime/llm.py

@@ -19,6 +19,7 @@ from pipecat.adapters.services.open_ai_realtime_adapter import (
     OpenAIRealtimeLLMAdapter,
 )
 from pipecat.frames.frames import (
+    AggregationType,
     BotStoppedSpeakingFrame,
     CancelFrame,
     EndFrame,
@@ -684,7 +685,7 @@ class OpenAIRealtimeLLMService(LLMService):
         # We receive audio transcript deltas (as opposed to text deltas) when
         # the output modality is "audio" (the default)
         if evt.delta:
-            frame = TTSTextFrame(evt.delta)
+            frame = TTSTextFrame(evt.delta, aggregated_by=AggregationType.SENTENCE)
             # OpenAI Realtime text already includes any necessary inter-chunk spaces
             frame.includes_inter_frame_spaces = True
             await self.push_frame(frame)

src/pipecat/services/openai_realtime_beta/openai.py

@@ -17,6 +17,7 @@ from loguru import logger
 
 from pipecat.adapters.services.open_ai_realtime_adapter import OpenAIRealtimeLLMAdapter
 from pipecat.frames.frames import (
+    AggregationType,
     BotStoppedSpeakingFrame,
     CancelFrame,
     EndFrame,
@@ -652,7 +653,7 @@ class OpenAIRealtimeBetaLLMService(LLMService):
     async def _handle_evt_audio_transcript_delta(self, evt):
         if evt.delta:
             await self.push_frame(LLMTextFrame(evt.delta))
-            await self.push_frame(TTSTextFrame(evt.delta))
+            await self.push_frame(TTSTextFrame(evt.delta, aggregated_by=AggregationType.SENTENCE))
 
     async def _handle_evt_speech_started(self, evt):
         await self._truncate_current_audio_response()

src/pipecat/services/rime/tts.py

@@ -113,6 +113,10 @@ class RimeTTSService(AudioContextWordTTSService):
             sample_rate: Audio sample rate in Hz.
             params: Additional configuration parameters.
             text_aggregator: Custom text aggregator for processing input text.
+
+                .. deprecated:: 0.0.95
+                    Use an LLMTextProcessor before the TTSService for custom text aggregation.
+
             aggregate_sentences: Whether to aggregate sentences within the TTSService.
             **kwargs: Additional arguments passed to parent class.
         """
@@ -123,10 +127,17 @@ class RimeTTSService(AudioContextWordTTSService):
             push_stop_frames=True,
             pause_frame_processing=True,
             sample_rate=sample_rate,
-            text_aggregator=text_aggregator or SkipTagsAggregator([("spell(", ")")]),
             **kwargs,
         )
 
+        if not text_aggregator:
+            # Always skip tags added for spelled-out text
+            # Note: This is primarily to support backwards compatibility.
+            #    The preferred way of taking advantage of Rime spelling is
+            #    to use an LLMTextProcessor and/or a text_transformer to identify
+            #    and insert these tags for the purpose of the TTS service alone.
+            self._text_aggregator = SkipTagsAggregator([("spell(", ")")])
+
         params = params or RimeTTSService.InputParams()
 
         # Store service configuration
@@ -152,6 +163,7 @@ class RimeTTSService(AudioContextWordTTSService):
         self._context_id = None  # Tracks current turn
         self._receive_task = None
         self._cumulative_time = 0  # Accumulates time across messages
+        self._extra_msg_fields = {}  # Extra fields for next message
 
     def can_generate_metrics(self) -> bool:
         """Check if this service can generate processing metrics.
@@ -181,6 +193,31 @@ class RimeTTSService(AudioContextWordTTSService):
         self._model = model
         await super().set_model(model)
 
+    # A set of Rime-specific helpers for text transformations
+    def SPELL(text: str) -> str:
+        """Wrap text in Rime spell function."""
+        return f"spell({text})"
+
+    def PAUSE_TAG(seconds: float) -> str:
+        """Convenience method to create a pause tag."""
+        return f"<{seconds * 1000}>"
+
+    def PRONOUNCE(self, text: str, word: str, phoneme: str) -> str:
+        """Convenience method to support Rime's custom pronunciations feature.
+
+        https://docs.rime.ai/api-reference/custom-pronunciation
+        """
+        self._extra_msg_fields["phonemizeBetweenBrackets"] = True
+        return text.replace(word, f"{phoneme}")
+
+    def INLINE_SPEED(self, text: str, speed: float) -> str:
+        """Convenience method to support inline speeds."""
+        if not self._extra_msg_fields:
+            self._extra_msg_fields = {}
+        speed_vals = self._extra_msg_fields.get("inlineSpeedAlpha", "").split(",")
+        self._extra_msg_fields["inlineSpeedAlpha"] = ",".join(speed_vals + [str(speed)])
+        return f"[{text}]"
+
     async def _update_settings(self, settings: Mapping[str, Any]):
         """Update service settings and reconnect if voice changed."""
         prev_voice = self._voice_id
@@ -193,7 +230,11 @@ class RimeTTSService(AudioContextWordTTSService):
 
     def _build_msg(self, text: str = "") -> dict:
         """Build JSON message for Rime API."""
-        return {"text": text, "contextId": self._context_id}
+        msg = {"text": text, "contextId": self._context_id}
+        if self._extra_msg_fields:
+            msg |= self._extra_msg_fields
+            self._extra_msg_fields = {}
+        return msg
 
     def _build_clear_msg(self) -> dict:
         """Build clear operation message."""

src/pipecat/services/tts_service.py

@@ -12,6 +12,8 @@ from typing import (
     Any,
     AsyncGenerator,
     AsyncIterator,
+    Awaitable,
+    Callable,
     Dict,
     List,
     Mapping,
@@ -23,6 +25,8 @@ from typing import (
 from loguru import logger
 
 from pipecat.frames.frames import (
+    AggregatedTextFrame,
+    AggregationType,
     BotStartedSpeakingFrame,
     BotStoppedSpeakingFrame,
     CancelFrame,
@@ -101,6 +105,16 @@ class TTSService(AIService):
         sample_rate: Optional[int] = None,
         # Text aggregator to aggregate incoming tokens and decide when to push to the TTS.
         text_aggregator: Optional[BaseTextAggregator] = None,
+        # Types of text aggregations that should not be spoken.
+        skip_aggregator_types: Optional[List[str]] = [],
+        # A list of callables to transform text before just before sending it to TTS.
+        # Each callable takes the aggregated text and its type, and returns the transformed text.
+        # To register, provide a list of tuples of (aggregation_type | '*', transform_function).
+        text_transforms: Optional[
+            List[
+                Tuple[AggregationType | str, Callable[[str, str | AggregationType], Awaitable[str]]]
+            ]
+        ] = None,
         # Text filter executed after text has been aggregated.
         text_filters: Optional[Sequence[BaseTextFilter]] = None,
         text_filter: Optional[BaseTextFilter] = None,
@@ -120,6 +134,16 @@ class TTSService(AIService):
             pause_frame_processing: Whether to pause frame processing during audio generation.
             sample_rate: Output sample rate for generated audio.
             text_aggregator: Custom text aggregator for processing incoming text.
+
+                .. deprecated:: 0.0.95
+                    Use an LLMTextProcessor before the TTSService for custom text aggregation.
+
+            skip_aggregator_types: List of aggregation types that should not be spoken.
+            text_transforms: A list of callables to transform text before just before sending it
+                to TTS. Each callable takes the aggregated text and its type, and returns the
+                transformed text. To register, provide a list of tuples of
+                (aggregation_type | '*', transform_function).
+
             text_filters: Sequence of text filters to apply after aggregation.
             text_filter: Single text filter (deprecated, use text_filters).
 
@@ -142,7 +166,21 @@ class TTSService(AIService):
         self._voice_id: str = ""
         self._settings: Dict[str, Any] = {}
         self._text_aggregator: BaseTextAggregator = text_aggregator or SimpleTextAggregator()
-        self._aggregated_text_includes_inter_frame_spaces: bool = False
+        if text_aggregator:
+            import warnings
+
+            with warnings.catch_warnings():
+                warnings.simplefilter("always")
+                warnings.warn(
+                    "Parameter 'text_aggregator' is deprecated. Use an LLMTextProcessor before the TTSService for custom text aggregation.",
+                    DeprecationWarning,
+                )
+
+        self._skip_aggregator_types: List[str] = skip_aggregator_types or []
+        self._text_transforms: List[
+            Tuple[AggregationType | str, Callable[[str, AggregationType | str], Awaitable[str]]]
+        ] = text_transforms or []
+        # TODO: Deprecate _text_filters when added to LLMTextProcessor
         self._text_filters: Sequence[BaseTextFilter] = text_filters or []
         self._transport_destination: Optional[str] = transport_destination
         self._tracing_enabled: bool = False
@@ -282,6 +320,39 @@ class TTSService(AIService):
             await self.cancel_task(self._stop_frame_task)
             self._stop_frame_task = None
 
+    def add_text_transformer(
+        self,
+        transform_function: Callable[[str, AggregationType | str], Awaitable[str]],
+        aggregation_type: AggregationType | str = "*",
+    ):
+        """Transform text for a specific aggregation type.
+
+        Args:
+            transform_function: The function to apply for transformation. This function should take
+                the text and aggregation type as input and return the transformed text.
+                Ex.: async def my_transform(text: str, aggregation_type: str) -> str:
+            aggregation_type: The type of aggregation to transform. This value defaults to "*" indicating
+                the function should handle all text before sending to TTS.
+        """
+        self._text_transforms.append((aggregation_type, transform_function))
+
+    def remove_text_transformer(
+        self,
+        transform_function: Callable[[str, AggregationType | str], Awaitable[str]],
+        aggregation_type: AggregationType | str = "*",
+    ):
+        """Remove a text transformer for a specific aggregation type.
+
+        Args:
+            transform_function: The function to remove.
+            aggregation_type: The type of aggregation to remove the transformer for.
+        """
+        self._text_transforms = [
+            (agg_type, func)
+            for agg_type, func in self._text_transforms
+            if not (agg_type == aggregation_type and func == transform_function)
+        ]
+
     async def _update_settings(self, settings: Mapping[str, Any]):
         for key, value in settings.items():
             if key in self._settings:
@@ -337,6 +408,8 @@ class TTSService(AIService):
             and frame.skip_tts
         ):
             await self.push_frame(frame, direction)
+        elif isinstance(frame, AggregatedTextFrame):
+            await self._push_tts_frames(frame)
         elif (
             isinstance(frame, TextFrame)
             and not isinstance(frame, InterimTranscriptionFrame)
@@ -352,17 +425,16 @@ class TTSService(AIService):
             # pause to avoid audio overlapping.
             await self._maybe_pause_frame_processing()
 
-            sentence = self._text_aggregator.text
+            pending_aggregation = self._text_aggregator.text
-            includes_inter_frame_spaces = self._aggregated_text_includes_inter_frame_spaces
 
             # Reset aggregator state
             await self._text_aggregator.reset()
             self._processing_text = False
-            self._aggregated_text_includes_inter_frame_spaces = False
 
-            await self._push_tts_frames(
+            if pending_aggregation.text:
-                sentence, includes_inter_frame_spaces=includes_inter_frame_spaces
+                await self._push_tts_frames(
-            )
+                    AggregatedTextFrame(pending_aggregation.text, pending_aggregation.type)
+                )
             if isinstance(frame, LLMFullResponseEndFrame):
                 if self._push_text_frames:
                     await self.push_frame(frame, direction)
@@ -372,7 +444,7 @@ class TTSService(AIService):
             # Store if we were processing text or not so we can set it back.
             processing_text = self._processing_text
             # Assumption: text in TTSSpeakFrame does not include inter-frame spaces
-            await self._push_tts_frames(frame.text, includes_inter_frame_spaces=False)
+            await self._push_tts_frames(AggregatedTextFrame(frame.text, AggregationType.SENTENCE))
             # We pause processing incoming frames because we are sending data to
             # the TTS. We pause to avoid audio overlapping.
             await self._maybe_pause_frame_processing()
@@ -462,21 +534,35 @@ class TTSService(AIService):
 
     async def _process_text_frame(self, frame: TextFrame):
         text: Optional[str] = None
+        includes_inter_frame_spaces: bool = False
         if not self._aggregate_sentences:
             text = frame.text
+            includes_inter_frame_spaces = frame.includes_inter_frame_spaces
+            aggregated_by = "token"
         else:
-            text = await self._text_aggregator.aggregate(frame.text)
+            aggregate = await self._text_aggregator.aggregate(frame.text)
-            # Assumption: whether inter-frame spaces are included shouldn't
+            if aggregate:
-            # change during aggregation, so we can just use the latest frame's
+                text = aggregate.text
-            # value
+                aggregated_by = aggregate.type
-            self._aggregated_text_includes_inter_frame_spaces = frame.includes_inter_frame_spaces
 
         if text:
+            logger.trace(f"Pushing TTS frames for text: {text}, {aggregated_by}")
             await self._push_tts_frames(
-                text, includes_inter_frame_spaces=frame.includes_inter_frame_spaces
+                AggregatedTextFrame(text, aggregated_by), includes_inter_frame_spaces
             )
 
-    async def _push_tts_frames(self, text: str, includes_inter_frame_spaces: bool):
+    async def _push_tts_frames(
+        self, src_frame: AggregatedTextFrame, includes_inter_frame_spaces: Optional[bool] = False
+    ):
+        type = src_frame.aggregated_by
+        text = src_frame.text
+
+        # Skip sending to TTS if the aggregation type is in the skip list. Simply
+        # push the original frame downstream.
+        if type in self._skip_aggregator_types:
+            await self.push_frame(src_frame)
+            return
+
         # Remove leading newlines only
         text = text.lstrip("\n")
 
@@ -492,20 +578,44 @@ class TTSService(AIService):
 
         await self.start_processing_metrics()
 
-        # Process all filter.
+        # Process all filters.
         for filter in self._text_filters:
             await filter.reset_interruption()
             text = await filter.filter(text)
 
-        if text:
+        if not text.strip():
-            await self.process_generator(self.run_tts(text))
+            await self.stop_processing_metrics()
+            return
+
+        # To support use cases that may want to know the text before it's spoken, we
+        # push the AggregatedTextFrame version before transforming and sending to TTS.
+        # However, we do not want to add this text to the assistant context until it
+        # is spoken, so we set append_to_context to False.
+        src_frame.append_to_context = False
+        await self.push_frame(src_frame)
+
+        # Note: Text transformations are meant to only affect the text sent to the TTS for
+        # TTS-specific purposes. This allows for explicit TTS modifications (e.g., inserting
+        # TTS supported tags for spelling or emotion or replacing an @ with "at"). For TTS
+        # services that support word-level timestamps, this CAN affect the resulting context
+        # since the TTSTextFrames are generated from the TTS output stream
+        transformed_text = text
+        for aggregation_type, transform in self._text_transforms:
+            if aggregation_type == type or aggregation_type == "*":
+                transformed_text = await transform(transformed_text, type)
+        await self.process_generator(self.run_tts(transformed_text))
 
         await self.stop_processing_metrics()
 
         if self._push_text_frames:
-            # We send the original text after the audio. This way, if we are
+            # In TTS services that support word timestamps, the TTSTextFrames
-            # interrupted, the text is not added to the assistant context.
+            # are pushed as words are spoken. However, in the case where the TTS service
-            frame = TTSTextFrame(text)
+            # does not support word timestamps (i.e. _push_text_frames is True), we send
+            # the original (non-transformed) text after the TTS generation has completed.
+            # This way, if we are interrupted, the text is not added to the assistant
+            # context and the context that IS added does not include TTS-specific tags
+            # or transformations.
+            frame = TTSTextFrame(text, aggregated_by=type)
             frame.includes_inter_frame_spaces = includes_inter_frame_spaces
             await self.push_frame(frame)
 
@@ -635,7 +745,7 @@ class WordTTSService(TTSService):
             else:
                 # Assumption: word-by-word text frames don't include spaces, so
                 # we can rely on the default includes_inter_frame_spaces=False
-                frame = TTSTextFrame(word)
+                frame = TTSTextFrame(word, aggregated_by=AggregationType.WORD)
                 frame.pts = self._initial_word_timestamp + timestamp
             if frame:
                 last_pts = frame.pts

src/pipecat/tests/utils.py

@@ -203,8 +203,16 @@ async def run_test(
             if not isinstance(frame, EndFrame) or not send_end_frame:
                 received_down_frames.append(frame)
 
-        print("received DOWN frames =", received_down_frames)
+        down_frames_printed = "["
-        print("expected DOWN frames =", expected_down_frames)
+        for frame in received_down_frames:
+            down_frames_printed += f"{frame.__class__.__name__}, "
+        down_frames_printed += "]"
+        expected_frames_printed = "["
+        for frame in expected_down_frames:
+            expected_frames_printed += f"{frame.__name__}, "
+        expected_frames_printed += "]"
+        print("received DOWN frames =", down_frames_printed)
+        print("expected DOWN frames =", expected_frames_printed)
 
         assert len(received_down_frames) == len(expected_down_frames)
 

src/pipecat/utils/text/base_text_aggregator.py

@@ -12,9 +12,47 @@ aggregated text should be sent for speech synthesis.
 """
 
 from abc import ABC, abstractmethod
+from dataclasses import dataclass
+from enum import Enum
 from typing import Optional
 
 
+class AggregationType(str, Enum):
+    """Built-in aggregation strings."""
+
+    SENTENCE = "sentence"
+    WORD = "word"
+
+    def __str__(self):
+        return self.value
+
+
+@dataclass
+class Aggregation:
+    """Data class representing aggregated text and its type.
+
+    An Aggregation object is created whenever a stream of text is aggregated by
+    a text aggregator. It contains the aggregated text and a type indicating
+    the nature of the aggregation.
+
+    Parameters:
+        text: The aggregated text content.
+        type: The type of aggregation the text represents (e.g., 'sentence', 'word', 'token',
+              'my_custom_aggregation').
+    """
+
+    text: str
+    type: str
+
+    def __str__(self) -> str:
+        """Return a string representation of the aggregation.
+
+        Returns:
+            A descriptive string showing the type and text of the aggregation.
+        """
+        return f"Aggregation by {self.type}: {self.text}"
+
+
 class BaseTextAggregator(ABC):
     """Base class for text aggregators in the Pipecat framework.
 
@@ -30,7 +68,7 @@ class BaseTextAggregator(ABC):
 
     @property
     @abstractmethod
-    def text(self) -> str:
+    def text(self) -> Aggregation:
         """Get the currently aggregated text.
 
         Subclasses must implement this property to return the text that has
@@ -42,25 +80,33 @@ class BaseTextAggregator(ABC):
         pass
 
     @abstractmethod
-    async def aggregate(self, text: str) -> Optional[str]:
+    async def aggregate(self, text: str) -> Optional[Aggregation]:
         """Aggregate the specified text with the currently accumulated text.
 
         This method should be implemented to define how the new text contributes
-        to the aggregation process. It returns the updated aggregated text if
+        to the aggregation process. It returns the aggregated text and a string
-        it's ready to be processed, or None otherwise.
+        describing how it was aggregated if it's ready to be processed,
+        or None otherwise.
 
         Subclasses should implement their specific logic for:
 
         - How to combine new text with existing accumulated text
         - When to consider the aggregated text ready for processing
         - What criteria determine text completion (e.g., sentence boundaries)
+        - When a completion occurs, the method should return an Aggregation object
+          containing the aggregated text and its type. The text should be stripped
+          of leading/trailing whitespace so that consumers can rely on a consistent
+          format.
 
         Args:
             text: The text to be aggregated.
 
         Returns:
-            The updated aggregated text if ready for processing, or None if more
+            An Aggregation object if ready for processing, or None if more
-            text is needed before the aggregated content is ready.
+            text is needed before the aggregated content is ready. If an Aggregation
+            object is returned, it should consist of the updated aggregated text,
+            stripped of leading/trailing whitespace, and a string indicating the
+            type of aggregation (e.g., 'sentence', 'word', 'token', 'my_custom_aggregation').
         """
         pass
 

src/pipecat/utils/text/pattern_pair_aggregator.py

@@ -8,19 +8,41 @@
 
 This module provides an aggregator that identifies and processes content between
 pattern pairs (like XML tags or custom delimiters) in streaming text, with
-support for custom handlers and configurable pattern removal.
+support for custom handlers and configurable actions for when a pattern is found.
 """
 
 import re
-from typing import Awaitable, Callable, Optional, Tuple
+from enum import Enum
+from typing import Awaitable, Callable, List, Optional, Tuple
 
 from loguru import logger
 
 from pipecat.utils.string import match_endofsentence
-from pipecat.utils.text.base_text_aggregator import BaseTextAggregator
+from pipecat.utils.text.base_text_aggregator import Aggregation, AggregationType, BaseTextAggregator
 
 
-class PatternMatch:
+class MatchAction(Enum):
+    """Actions to take when a pattern pair is matched.
+
+    Parameters:
+        REMOVE: The text along with its delimiters will be removed from the streaming text.
+              Sentence aggregation will continue on as if this text did not exist.
+        KEEP: The delimiters will be removed, but the content between them will be kept.
+              Sentence aggregation will continue on with the internal text included.
+        AGGREGATE: The delimiters will be removed and the content between will be treated
+              as a separate aggregation. Any text before the start of the pattern will be
+              returned early, whether or not a complete sentence was found. Then the pattern
+              will be returned. Then the aggregation will continue on sentence matching after
+              the closing delimiter is found. The content between the delimiters is not
+              aggregated by sentence. It is aggregated as one single block of text.
+    """
+
+    REMOVE = "remove"
+    KEEP = "keep"
+    AGGREGATE = "aggregate"
+
+
+class PatternMatch(Aggregation):
     """Represents a matched pattern pair with its content.
 
     A PatternMatch object is created when a complete pattern pair is found
@@ -29,25 +51,25 @@ class PatternMatch:
     content between the patterns.
     """
 
-    def __init__(self, pattern_id: str, full_match: str, content: str):
+    def __init__(self, content: str, type: str, full_match: str):
         """Initialize a pattern match.
 
         Args:
-            pattern_id: The identifier of the matched pattern pair.
+            type: The type of the matched pattern pair. It should be representative
+                   of the content type (e.g., 'sentence', 'code', 'speaker', 'custom').
             full_match: The complete text including start and end patterns.
             content: The text content between the start and end patterns.
         """
-        self.pattern_id = pattern_id
+        super().__init__(text=content, type=type)
         self.full_match = full_match
-        self.content = content
 
     def __str__(self) -> str:
         """Return a string representation of the pattern match.
 
         Returns:
-            A descriptive string showing the pattern ID and content.
+            A descriptive string showing the pattern type and content.
         """
-        return f"PatternMatch(id={self.pattern_id}, content={self.content})"
+        return f"PatternMatch(type={self.type}, text={self.text}, full_match={self.full_match})"
 
 
 class PatternPairAggregator(BaseTextAggregator):
@@ -55,16 +77,21 @@ class PatternPairAggregator(BaseTextAggregator):
 
     This aggregator buffers text until it can identify complete pattern pairs
     (defined by start and end patterns), processes the content between these
-    patterns using registered handlers, and returns text at sentence boundaries.
+    patterns using registered handlers. By default, its aggregation method
-    It's particularly useful for processing structured content in streaming text,
+    returns text at sentence boundaries, and remove the content found between
-    such as XML tags, markdown formatting, or custom delimiters.
+    any matched patterns. However, matched patterns can also be configured to
+    returned as a separate aggregation object containing the content between
+    their start and end patterns or left in, so that only the delimiters are
+    removed and a callback can be triggered.
+
+    This aggregator is particularly useful for processing structured content in
+    streaming text, such as XML tags, markdown formatting, or custom delimiters.
 
     The aggregator ensures that patterns spanning multiple text chunks are
-    correctly identified and handles cases where patterns contain sentence
+    correctly identified.
-    boundaries.
     """
 
-    def __init__(self):
+    def __init__(self, **kwargs):
         """Initialize the pattern pair aggregator.
 
         Creates an empty aggregator with no patterns or handlers registered.
@@ -75,16 +102,27 @@ class PatternPairAggregator(BaseTextAggregator):
         self._handlers = {}
 
     @property
-    def text(self) -> str:
+    def text(self) -> Aggregation:
-        """Get the currently buffered text.
+        """Get the currently aggregated text.
 
         Returns:
-            The current text buffer content that hasn't been processed yet.
+            The text that has been accumulated in the buffer.
         """
-        return self._text
+        pattern_start = self._match_start_of_pattern(self._text)
+        stripped_text = self._text.strip()
+        type = (
+            pattern_start[1].get("type", AggregationType.SENTENCE)
+            if pattern_start
+            else AggregationType.SENTENCE
+        )
+        return Aggregation(text=stripped_text, type=type)
 
-    def add_pattern_pair(
+    def add_pattern(
-        self, pattern_id: str, start_pattern: str, end_pattern: str, remove_match: bool = True
+        self,
+        type: str,
+        start_pattern: str,
+        end_pattern: str,
+        action: MatchAction = MatchAction.REMOVE,
     ) -> "PatternPairAggregator":
         """Add a pattern pair to detect in the text.
 
@@ -93,41 +131,94 @@ class PatternPairAggregator(BaseTextAggregator):
         the end pattern, and treat the content between them as a match.
 
         Args:
-            pattern_id: Unique identifier for this pattern pair.
+            type: Identifier for this pattern pair. Should be unique and ideally descriptive.
+                  (e.g., 'code', 'speaker', 'custom'). type can not be 'sentence' or 'word' as
+                  those are reserved for the default behavior.
             start_pattern: Pattern that marks the beginning of content.
             end_pattern: Pattern that marks the end of content.
-            remove_match: Whether to remove the matched content from the text.
+            action: What to do when a complete pattern is matched:
+                    - MatchAction.REMOVE: Remove the matched pattern from the text.
+                    - MatchAction.KEEP: Keep the matched pattern in the text and treat it as
+                                        normal text. This allows you to register handlers for
+                                        the pattern without affecting the aggregation logic.
+                    - MatchAction.AGGREGATE: Return the matched pattern as a separate
+                                             aggregation object.
 
         Returns:
             Self for method chaining.
         """
-        self._patterns[pattern_id] = {
+        if type in [AggregationType.SENTENCE, AggregationType.WORD]:
+            raise ValueError(
+                f"The aggregation type '{type}' is reserved for default behavior and can not be used for custom patterns."
+            )
+        self._patterns[type] = {
             "start": start_pattern,
             "end": end_pattern,
-            "remove_match": remove_match,
+            "type": type,
+            "action": action,
         }
         return self
 
+    def add_pattern_pair(
+        self, pattern_id: str, start_pattern: str, end_pattern: str, remove_match: bool = True
+    ):
+        """Add a pattern pair to detect in the text.
+
+        .. deprecated:: 0.0.95
+            This function is deprecated and will be removed in a future version.
+            Use `add_pattern` with a type and MatchAction instead.
+
+            This method calls `add_pattern` setting type with the provided pattern_id and action
+            to either MatchAction.REMOVE or MatchAction.KEEP based on `remove_match`.
+
+        Args:
+            pattern_id: Identifier for this pattern pair. Should be unique and ideally descriptive.
+                        (e.g., 'code', 'speaker', 'custom'). pattern_id can not be 'sentence' or 'word'
+                        as those arereserved for the default behavior.
+            start_pattern: Pattern that marks the beginning of content.
+            end_pattern: Pattern that marks the end of content.
+            remove_match: If True, the matched pattern will be removed from the text. (Same as MatchAction.REMOVE)
+                          If False, it will be kept and treated as normal text. (Same as MatchAction.KEEP)
+        """
+        import warnings
+
+        with warnings.catch_warnings():
+            warnings.simplefilter("once")
+            warnings.warn(
+                "add_pattern_pair with a pattern_id or remove_match is deprecated and will be"
+                " removed in a future version. Use add_pattern with a type and MatchAction instead",
+                DeprecationWarning,
+                stacklevel=2,
+            )
+
+        action = MatchAction.REMOVE if remove_match else MatchAction.KEEP
+        return self.add_pattern(
+            type=pattern_id,
+            start_pattern=start_pattern,
+            end_pattern=end_pattern,
+            action=action,
+        )
+
     def on_pattern_match(
-        self, pattern_id: str, handler: Callable[[PatternMatch], Awaitable[None]]
+        self, type: str, handler: Callable[[PatternMatch], Awaitable[None]]
     ) -> "PatternPairAggregator":
         """Register a handler for when a pattern pair is matched.
 
         The handler will be called whenever a complete match for the
-        specified pattern ID is found in the text.
+        specified type is found in the text.
 
         Args:
-            pattern_id: ID of the pattern pair to match.
+            type: The type of the pattern pair to trigger the handler.
             handler: Async function to call when pattern is matched.
                      The function should accept a PatternMatch object.
 
         Returns:
             Self for method chaining.
         """
-        self._handlers[pattern_id] = handler
+        self._handlers[type] = handler
         return self
 
-    async def _process_complete_patterns(self, text: str) -> Tuple[str, bool]:
+    async def _process_complete_patterns(self, text: str) -> Tuple[List[PatternMatch], str]:
         """Process all complete pattern pairs in the text.
 
         Searches for all complete pattern pairs in the text, calls the
@@ -137,19 +228,19 @@ class PatternPairAggregator(BaseTextAggregator):
             text: The text to process for pattern matches.
 
         Returns:
-            Tuple of (processed_text, was_modified) where:
+            Tuple of (all_matches, processed_text) where:
 
-            - processed_text is the text after processing patterns
+            - all_matches is a list of all pattern matches found. Note: There really should only ever be 1.
-            - was_modified indicates whether any changes were made
+            - processed_text is the text after processing patterns. If no patterns are found, it will be the same as input text.
         """
+        all_matches = []
         processed_text = text
-        modified = False
 
-        for pattern_id, pattern_info in self._patterns.items():
+        for type, pattern_info in self._patterns.items():
             # Escape special regex characters in the patterns
             start = re.escape(pattern_info["start"])
             end = re.escape(pattern_info["end"])
-            remove_match = pattern_info["remove_match"]
+            action = pattern_info["action"]
 
             # Create regex to match from start pattern to end pattern
             # The .*? is non-greedy to handle nested patterns
@@ -165,24 +256,25 @@ class PatternPairAggregator(BaseTextAggregator):
 
                 # Create pattern match object
                 pattern_match = PatternMatch(
-                    pattern_id=pattern_id, full_match=full_match, content=content
+                    content=content.strip(), type=type, full_match=full_match
                 )
 
                 # Call the appropriate handler if registered
-                if pattern_id in self._handlers:
+                if type in self._handlers:
                     try:
-                        await self._handlers[pattern_id](pattern_match)
+                        await self._handlers[type](pattern_match)
                     except Exception as e:
-                        logger.error(f"Error in pattern handler for {pattern_id}: {e}")
+                        logger.error(f"Error in pattern handler for {type}: {e}")
 
                 # Remove the pattern from the text if configured
-                if remove_match:
+                if action == MatchAction.REMOVE:
                     processed_text = processed_text.replace(full_match, "", 1)
-                    modified = True
+                else:
+                    all_matches.append(pattern_match)
 
-        return processed_text, modified
+        return all_matches, processed_text
 
-    def _has_incomplete_patterns(self, text: str) -> bool:
+    def _match_start_of_pattern(self, text: str) -> Optional[Tuple[int, dict]]:
         """Check if text contains incomplete pattern pairs.
 
         Determines whether the text contains any start patterns without
@@ -192,9 +284,10 @@ class PatternPairAggregator(BaseTextAggregator):
             text: The text to check for incomplete patterns.
 
         Returns:
-            True if there are incomplete patterns, False otherwise.
+            A tuple of (start_index, pattern_info) if an incomplete pattern is found,
+            or None if no patterns are found or all patterns are complete.
         """
-        for pattern_id, pattern_info in self._patterns.items():
+        for type, pattern_info in self._patterns.items():
             start = pattern_info["start"]
             end = pattern_info["end"]
 
@@ -203,12 +296,16 @@ class PatternPairAggregator(BaseTextAggregator):
             end_count = text.count(end)
 
             # If there are more starts than ends, we have incomplete patterns
+            # Again, this is written generically but there only ever should
+            # be one pattern active at a time, so the counts should be 0 or 1.
+            # Which is why we base the return on the first found.
             if start_count > end_count:
-                return True
+                start_index = text.find(start)
+                return [start_index, pattern_info]
 
-        return False
+        return None
 
-    async def aggregate(self, text: str) -> Optional[str]:
+    async def aggregate(self, text: str) -> Optional[PatternMatch]:
         """Aggregate text and process pattern pairs.
 
         This method adds the new text to the buffer, processes any complete pattern
@@ -227,16 +324,36 @@ class PatternPairAggregator(BaseTextAggregator):
         self._text += text
 
         # Process any complete patterns in the buffer
-        processed_text, modified = await self._process_complete_patterns(self._text)
+        patterns, processed_text = await self._process_complete_patterns(self._text)
 
-        # Only update the buffer if modifications were made
+        self._text = processed_text
-        if modified:
+
-            self._text = processed_text
+        if len(patterns) > 0:
+            if len(patterns) > 1:
+                logger.warning(
+                    f"Multiple patterns matched: {[p.type for p in patterns]}. Only the first pattern will be returned."
+                )
+            # If the pattern found is set to be aggregated, return it
+            action = self._patterns[patterns[0].type].get("action", MatchAction.REMOVE)
+            if action == MatchAction.AGGREGATE:
+                self._text = ""
+                return patterns[0]
 
         # Check if we have incomplete patterns
-        if self._has_incomplete_patterns(self._text):
+        pattern_start = self._match_start_of_pattern(self._text)
-            # Still waiting for complete patterns
+        if pattern_start is not None:
-            return None
+            # If the start pattern is at the beginning or should not be separately aggregated, return None
+            if (
+                pattern_start[0] == 0
+                or pattern_start[1].get("action", MatchAction.REMOVE) != MatchAction.AGGREGATE
+            ):
+                return None
+            # Otherwise, strip the text up to the start pattern and return it
+            result = self._text[: pattern_start[0]]
+            self._text = self._text[pattern_start[0] :]
+            return PatternMatch(
+                content=result.strip(), type=AggregationType.SENTENCE, full_match=result
+            )
 
         # Find sentence boundary if no incomplete patterns
         eos_marker = match_endofsentence(self._text)
@@ -244,7 +361,9 @@ class PatternPairAggregator(BaseTextAggregator):
             # Extract text up to the sentence boundary
             result = self._text[:eos_marker]
             self._text = self._text[eos_marker:]
-            return result
+            return PatternMatch(
+                content=result.strip(), type=AggregationType.SENTENCE, full_match=result
+            )
 
         # No complete sentence found yet
         return None

src/pipecat/utils/text/simple_text_aggregator.py

@@ -14,7 +14,7 @@ text processing scenarios.
 from typing import Optional
 
 from pipecat.utils.string import match_endofsentence
-from pipecat.utils.text.base_text_aggregator import BaseTextAggregator
+from pipecat.utils.text.base_text_aggregator import Aggregation, AggregationType, BaseTextAggregator
 
 
 class SimpleTextAggregator(BaseTextAggregator):
@@ -33,15 +33,15 @@ class SimpleTextAggregator(BaseTextAggregator):
         self._text = ""
 
     @property
-    def text(self) -> str:
+    def text(self) -> Aggregation:
         """Get the currently aggregated text.
 
         Returns:
             The text that has been accumulated in the buffer.
         """
-        return self._text
+        return Aggregation(text=self._text.strip(), type=AggregationType.SENTENCE)
 
-    async def aggregate(self, text: str) -> Optional[str]:
+    async def aggregate(self, text: str) -> Optional[Aggregation]:
         """Aggregate text and return completed sentences.
 
         Adds the new text to the buffer and checks for end-of-sentence markers.
@@ -64,7 +64,9 @@ class SimpleTextAggregator(BaseTextAggregator):
             result = self._text[:eos_end_marker]
             self._text = self._text[eos_end_marker:]
 
-        return result
+        if result:
+            return Aggregation(text=result.strip(), type=AggregationType.SENTENCE)
+        return None
 
     async def handle_interruption(self):
         """Handle interruptions by clearing the text buffer.

src/pipecat/utils/text/skip_tags_aggregator.py

@@ -14,7 +14,7 @@ as a unit regardless of internal punctuation.
 from typing import Optional, Sequence
 
 from pipecat.utils.string import StartEndTags, match_endofsentence, parse_start_end_tags
-from pipecat.utils.text.base_text_aggregator import BaseTextAggregator
+from pipecat.utils.text.base_text_aggregator import Aggregation, AggregationType, BaseTextAggregator
 
 
 class SkipTagsAggregator(BaseTextAggregator):
@@ -43,15 +43,15 @@ class SkipTagsAggregator(BaseTextAggregator):
         self._current_tag_index: int = 0
 
     @property
-    def text(self) -> str:
+    def text(self) -> Aggregation:
         """Get the currently buffered text.
 
         Returns:
             The current text buffer content that hasn't been processed yet.
         """
-        return self._text
+        return Aggregation(text=self._text.strip(), type=AggregationType.SENTENCE)
 
-    async def aggregate(self, text: str) -> Optional[str]:
+    async def aggregate(self, text: str) -> Optional[Aggregation]:
         """Aggregate text while respecting tag boundaries.
 
         This method adds the new text to the buffer, processes any complete
@@ -63,8 +63,9 @@ class SkipTagsAggregator(BaseTextAggregator):
             text: New text to add to the buffer.
 
         Returns:
-            Processed text up to a sentence boundary (when not within tags),
+            An Aggregation object containing text up to a sentence boundary and
-            or None if more text is needed to complete a sentence or close tags.
+            marked as SENTENCE type or None if more text is needed to complete a
+            sentence or close tags.
         """
         # Add new text to buffer
         self._text += text
@@ -80,7 +81,7 @@ class SkipTagsAggregator(BaseTextAggregator):
                 # Extract text up to the sentence boundary
                 result = self._text[:eos_marker]
                 self._text = self._text[eos_marker:]
-                return result
+                return Aggregation(text=result.strip(), type=AggregationType.SENTENCE)
 
         # No complete sentence found yet
         return None

tests/test_pattern_pair_aggregator.py

@@ -7,30 +7,42 @@
 import unittest
 from unittest.mock import AsyncMock
 
-from pipecat.utils.text.pattern_pair_aggregator import PatternMatch, PatternPairAggregator
+from pipecat.utils.text.pattern_pair_aggregator import (
+    MatchAction,
+    PatternMatch,
+    PatternPairAggregator,
+)
 
 
 class TestPatternPairAggregator(unittest.IsolatedAsyncioTestCase):
     def setUp(self):
         self.aggregator = PatternPairAggregator()
         self.test_handler = AsyncMock()
+        self.code_handler = AsyncMock()
 
         # Add a test pattern
         self.aggregator.add_pattern_pair(
             pattern_id="test_pattern",
             start_pattern="<test>",
             end_pattern="</test>",
-            remove_match=True,
+        )
+        self.aggregator.add_pattern(
+            type="code_pattern",
+            start_pattern="<code>",
+            end_pattern="</code>",
+            action=MatchAction.AGGREGATE,
         )
 
         # Register the mock handler
         self.aggregator.on_pattern_match("test_pattern", self.test_handler)
+        self.aggregator.on_pattern_match("code_pattern", self.code_handler)
 
     async def test_pattern_match_and_removal(self):
         # First part doesn't complete the pattern
         result = await self.aggregator.aggregate("Hello <test>pattern")
         self.assertIsNone(result)
-        self.assertEqual(self.aggregator.text, "Hello <test>pattern")
+        self.assertEqual(self.aggregator.text.text, "Hello <test>pattern")
+        self.assertEqual(self.aggregator.text.type, "test_pattern")
 
         # Second part completes the pattern and includes an exclamation point
         result = await self.aggregator.aggregate(" content</test>!")
@@ -39,20 +51,50 @@ class TestPatternPairAggregator(unittest.IsolatedAsyncioTestCase):
         self.test_handler.assert_called_once()
         call_args = self.test_handler.call_args[0][0]
         self.assertIsInstance(call_args, PatternMatch)
-        self.assertEqual(call_args.pattern_id, "test_pattern")
+        self.assertEqual(call_args.type, "test_pattern")
         self.assertEqual(call_args.full_match, "<test>pattern content</test>")
-        self.assertEqual(call_args.content, "pattern content")
+        self.assertEqual(call_args.text, "pattern content")
 
         # The exclamation point should be treated as a sentence boundary,
         # so the result should include just text up to and including "!"
-        self.assertEqual(result, "Hello !")
+        self.assertEqual(result.text, "Hello !")
+        self.assertEqual(result.type, "sentence")
+
+        # Next sentence should be processed separately. Spaces around the sentence
+        # should be stripped in the returned Aggregation.
+        result = await self.aggregator.aggregate(" This is another sentence.")
+        self.assertEqual(result.text, "This is another sentence.")
+
+        # Buffer should be empty after returning a complete sentence
+        self.assertEqual(self.aggregator.text.text, "")
+
+    async def test_pattern_match_and_aggregate(self):
+        # First part doesn't complete the pattern
+        result = await self.aggregator.aggregate("Here is code <code>pattern")
+        self.assertEqual(result.text, "Here is code")
+        self.assertEqual(self.aggregator.text.text, "<code>pattern")
+        self.assertEqual(self.aggregator.text.type, "code_pattern")
+
+        # Second part completes the pattern and includes an exclamation point
+        result = await self.aggregator.aggregate(" content</code>")
+
+        # Verify the handler was called with correct PatternMatch object
+        self.code_handler.assert_called_once()
+        call_args = self.code_handler.call_args[0][0]
+        self.assertIsInstance(call_args, PatternMatch)
+        self.assertEqual(call_args.type, "code_pattern")
+        self.assertEqual(call_args.full_match, "<code>pattern content</code>")
+        self.assertEqual(call_args.text, "pattern content")
+        self.assertEqual(result.text, "pattern content")
+        self.assertEqual(result.type, "code_pattern")
 
         # Next sentence should be processed separately
         result = await self.aggregator.aggregate(" This is another sentence.")
-        self.assertEqual(result, " This is another sentence.")
+        self.assertEqual(result.text, "This is another sentence.")
+        self.assertEqual(result.type, "sentence")
 
         # Buffer should be empty after returning a complete sentence
-        self.assertEqual(self.aggregator.text, "")
+        self.assertEqual(self.aggregator.text.text, "")
 
     async def test_incomplete_pattern(self):
         # Add text with incomplete pattern
@@ -65,26 +107,30 @@ class TestPatternPairAggregator(unittest.IsolatedAsyncioTestCase):
         self.test_handler.assert_not_called()
 
         # Buffer should contain the incomplete text
-        self.assertEqual(self.aggregator.text, "Hello <test>pattern content")
+        self.assertEqual(self.aggregator.text.text, "Hello <test>pattern content")
+        self.assertEqual(self.aggregator.text.type, "test_pattern")
 
         # Reset and confirm buffer is cleared
         await self.aggregator.reset()
-        self.assertEqual(self.aggregator.text, "")
+        self.assertEqual(self.aggregator.text.text, "")
 
     async def test_multiple_patterns(self):
         # Set up multiple patterns and handlers
         voice_handler = AsyncMock()
         emphasis_handler = AsyncMock()
 
-        self.aggregator.add_pattern_pair(
+        self.aggregator.add_pattern(
-            pattern_id="voice", start_pattern="<voice>", end_pattern="</voice>", remove_match=True
+            type="voice",
+            start_pattern="<voice>",
+            end_pattern="</voice>",
+            action=MatchAction.REMOVE,
         )
 
-        self.aggregator.add_pattern_pair(
+        self.aggregator.add_pattern(
-            pattern_id="emphasis",
+            type="emphasis",
             start_pattern="<em>",
             end_pattern="</em>",
-            remove_match=False,  # Keep emphasis tags
+            action=MatchAction.KEEP,  # Keep emphasis tags
         )
 
         self.aggregator.on_pattern_match("voice", voice_handler)
@@ -97,19 +143,19 @@ class TestPatternPairAggregator(unittest.IsolatedAsyncioTestCase):
         # Both handlers should be called with correct data
         voice_handler.assert_called_once()
         voice_match = voice_handler.call_args[0][0]
-        self.assertEqual(voice_match.pattern_id, "voice")
+        self.assertEqual(voice_match.type, "voice")
-        self.assertEqual(voice_match.content, "female")
+        self.assertEqual(voice_match.text, "female")
 
         emphasis_handler.assert_called_once()
         emphasis_match = emphasis_handler.call_args[0][0]
-        self.assertEqual(emphasis_match.pattern_id, "emphasis")
+        self.assertEqual(emphasis_match.type, "emphasis")
-        self.assertEqual(emphasis_match.content, "very")
+        self.assertEqual(emphasis_match.text, "very")
 
         # Voice pattern should be removed, emphasis pattern should remain
-        self.assertEqual(result, "Hello  I am <em>very</em> excited to meet you!")
+        self.assertEqual(result.text, "Hello  I am <em>very</em> excited to meet you!")
 
         # Buffer should be empty
-        self.assertEqual(self.aggregator.text, "")
+        self.assertEqual(self.aggregator.text.text, "")
 
     async def test_handle_interruption(self):
         # Start with incomplete pattern
@@ -120,7 +166,7 @@ class TestPatternPairAggregator(unittest.IsolatedAsyncioTestCase):
         await self.aggregator.handle_interruption()
 
         # Buffer should be cleared
-        self.assertEqual(self.aggregator.text, "")
+        self.assertEqual(self.aggregator.text.text, "")
 
         # Handler should not have been called
         self.test_handler.assert_not_called()
@@ -138,10 +184,10 @@ class TestPatternPairAggregator(unittest.IsolatedAsyncioTestCase):
         # Handler should be called with entire content
         self.test_handler.assert_called_once()
         call_args = self.test_handler.call_args[0][0]
-        self.assertEqual(call_args.content, "This is sentence one. This is sentence two.")
+        self.assertEqual(call_args.text, "This is sentence one. This is sentence two.")
 
         # Pattern should be removed, resulting in text with sentences merged
-        self.assertEqual(result, "Hello  Final sentence.")
+        self.assertEqual(result.text, "Hello  Final sentence.")
 
         # Buffer should be empty
-        self.assertEqual(self.aggregator.text, "")
+        self.assertEqual(self.aggregator.text.text, "")

tests/test_piper_tts.py

@@ -13,6 +13,7 @@ import pytest
 from aiohttp import web
 
 from pipecat.frames.frames import (
+    AggregatedTextFrame,
     ErrorFrame,
     TTSAudioRawFrame,
     TTSSpeakFrame,
@@ -74,6 +75,7 @@ async def test_run_piper_tts_success(aiohttp_client):
         ]
 
         expected_returned_frames = [
+            AggregatedTextFrame,
             TTSStartedFrame,
             TTSAudioRawFrame,
             TTSAudioRawFrame,
@@ -121,7 +123,7 @@ async def test_run_piper_tts_error(aiohttp_client):
             TTSSpeakFrame(text="Error case."),
         ]
 
-        expected_down_frames = [TTSStoppedFrame, TTSTextFrame]
+        expected_down_frames = [AggregatedTextFrame, TTSStoppedFrame, TTSTextFrame]
 
         expected_up_frames = [ErrorFrame]
 

tests/test_simple_text_aggregator.py

@@ -15,15 +15,21 @@ class TestSimpleTextAggregator(unittest.IsolatedAsyncioTestCase):
 
     async def test_reset_aggregations(self):
         assert await self.aggregator.aggregate("Hello ") == None
-        assert self.aggregator.text == "Hello "
+        assert self.aggregator.text.text == "Hello"
         await self.aggregator.reset()
-        assert self.aggregator.text == ""
+        assert self.aggregator.text.text == ""
 
     async def test_simple_sentence(self):
         assert await self.aggregator.aggregate("Hello ") == None
-        assert await self.aggregator.aggregate("Pipecat!") == "Hello Pipecat!"
+        aggregate = await self.aggregator.aggregate("Pipecat!")
-        assert self.aggregator.text == ""
+        assert aggregate.text == "Hello Pipecat!"
+        assert aggregate.type == "sentence"
+        assert self.aggregator.text.text == ""
 
     async def test_multiple_sentences(self):
-        assert await self.aggregator.aggregate("Hello Pipecat! How are ") == "Hello Pipecat!"
+        aggregate = await self.aggregator.aggregate("Hello Pipecat! How are ")
-        assert await self.aggregator.aggregate("you?") == " How are you?"
+        assert aggregate.text == "Hello Pipecat!"
+        # Aggregators should strip leading/trailing spaces when returning text
+        assert self.aggregator.text.text == "How are"
+        aggregate = await self.aggregator.aggregate("you?")
+        assert aggregate.text == "How are you?"

tests/test_skip_tags_aggregator.py

@@ -18,16 +18,18 @@ class TestSkipTagsAggregator(unittest.IsolatedAsyncioTestCase):
 
         # No tags involved, aggregate at end of sentence.
         result = await self.aggregator.aggregate("Hello Pipecat!")
-        self.assertEqual(result, "Hello Pipecat!")
+        self.assertEqual(result.text, "Hello Pipecat!")
-        self.assertEqual(self.aggregator.text, "")
+        self.assertEqual(result.type, "sentence")
+        self.assertEqual(self.aggregator.text.text, "")
 
     async def test_basic_tags(self):
         await self.aggregator.reset()
 
         # Tags involved, avoid aggregation during tags.
         result = await self.aggregator.aggregate("My email is <spell>foo@pipecat.ai</spell>.")
-        self.assertEqual(result, "My email is <spell>foo@pipecat.ai</spell>.")
+        self.assertEqual(result.text, "My email is <spell>foo@pipecat.ai</spell>.")
-        self.assertEqual(self.aggregator.text, "")
+        self.assertEqual(result.type, "sentence")
+        self.assertEqual(self.aggregator.text.text, "")
 
     async def test_streaming_tags(self):
         await self.aggregator.reset()
@@ -35,20 +37,22 @@ class TestSkipTagsAggregator(unittest.IsolatedAsyncioTestCase):
         # Tags involved, stream small chunk of texts.
         result = await self.aggregator.aggregate("My email is <sp")
         self.assertIsNone(result)
-        self.assertEqual(self.aggregator.text, "My email is <sp")
+        self.assertEqual(self.aggregator.text.text, "My email is <sp")
 
         result = await self.aggregator.aggregate("ell>foo.")
         self.assertIsNone(result)
-        self.assertEqual(self.aggregator.text, "My email is <spell>foo.")
+        self.assertEqual(self.aggregator.text.text, "My email is <spell>foo.")
 
         result = await self.aggregator.aggregate("bar@pipecat.")
         self.assertIsNone(result)
-        self.assertEqual(self.aggregator.text, "My email is <spell>foo.bar@pipecat.")
+        self.assertEqual(self.aggregator.text.text, "My email is <spell>foo.bar@pipecat.")
 
         result = await self.aggregator.aggregate("ai</spe")
         self.assertIsNone(result)
-        self.assertEqual(self.aggregator.text, "My email is <spell>foo.bar@pipecat.ai</spe")
+        self.assertEqual(self.aggregator.text.text, "My email is <spell>foo.bar@pipecat.ai</spe")
+        self.assertEqual(self.aggregator.text.type, "sentence")
 
         result = await self.aggregator.aggregate("ll>.")
-        self.assertEqual(result, "My email is <spell>foo.bar@pipecat.ai</spell>.")
+        self.assertEqual(result.text, "My email is <spell>foo.bar@pipecat.ai</spell>.")
-        self.assertEqual(self.aggregator.text, "")
+        self.assertEqual(self.aggregator.text.text, "")
+        self.assertEqual(self.aggregator.text.type, "sentence")

tests/test_transcript_processor.py

@@ -11,6 +11,7 @@ from datetime import datetime, timezone
 from typing import List, Tuple, cast
 
 from pipecat.frames.frames import (
+    AggregationType,
     BotStartedSpeakingFrame,
     BotStoppedSpeakingFrame,
     CancelFrame,
@@ -130,11 +131,11 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
         frames_to_send = [
             BotStartedSpeakingFrame(),
             SleepFrame(),  # Wait for StartedSpeaking to process
-            TTSTextFrame(text="Hello"),
+            TTSTextFrame(text="Hello", aggregated_by=AggregationType.WORD),
-            TTSTextFrame(text="world!"),
+            TTSTextFrame(text="world!", aggregated_by=AggregationType.WORD),
-            TTSTextFrame(text="How"),
+            TTSTextFrame(text="How", aggregated_by=AggregationType.WORD),
-            TTSTextFrame(text="are"),
+            TTSTextFrame(text="are", aggregated_by=AggregationType.WORD),
-            TTSTextFrame(text="you?"),
+            TTSTextFrame(text="you?", aggregated_by=AggregationType.WORD),
             SleepFrame(),  # Wait for text frames to queue
             BotStoppedSpeakingFrame(),
         ]
@@ -195,9 +196,9 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
         frames_to_send = [
             BotStartedSpeakingFrame(),
             SleepFrame(),
-            TTSTextFrame(text=""),  # Empty text
+            TTSTextFrame(text="", aggregated_by=AggregationType.WORD),  # Empty text
-            TTSTextFrame(text="   "),  # Just whitespace
+            TTSTextFrame(text="   ", aggregated_by=AggregationType.WORD),  # Just whitespace
-            TTSTextFrame(text="\n"),  # Just newline
+            TTSTextFrame(text="\n", aggregated_by=AggregationType.WORD),  # Just newline
             BotStoppedSpeakingFrame(),
             # Pipeline ends here; run_test will automatically send EndFrame
         ]
@@ -235,14 +236,14 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
         frames_to_send = [
             BotStartedSpeakingFrame(),
             SleepFrame(),
-            TTSTextFrame(text="Hello"),
+            TTSTextFrame(text="Hello", aggregated_by=AggregationType.WORD),
-            TTSTextFrame(text="world!"),
+            TTSTextFrame(text="world!", aggregated_by=AggregationType.WORD),
             SleepFrame(),
             InterruptionFrame(),  # User interrupts here
             SleepFrame(),
             BotStartedSpeakingFrame(),
-            TTSTextFrame(text="New"),
+            TTSTextFrame(text="New", aggregated_by=AggregationType.WORD),
-            TTSTextFrame(text="response"),
+            TTSTextFrame(text="response", aggregated_by=AggregationType.WORD),
             SleepFrame(),
             BotStoppedSpeakingFrame(),
         ]
@@ -299,8 +300,8 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
         frames_to_send = [
             BotStartedSpeakingFrame(),
             SleepFrame(),
-            TTSTextFrame(text="Hello"),
+            TTSTextFrame(text="Hello", aggregated_by=AggregationType.WORD),
-            TTSTextFrame(text="world"),
+            TTSTextFrame(text="world", aggregated_by=AggregationType.WORD),
             # Pipeline ends here; run_test will automatically send EndFrame
         ]
 
@@ -338,8 +339,8 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
         frames_to_send = [
             BotStartedSpeakingFrame(),
             SleepFrame(),
-            TTSTextFrame(text="Hello"),
+            TTSTextFrame(text="Hello", aggregated_by=AggregationType.WORD),
-            TTSTextFrame(text="world"),
+            TTSTextFrame(text="world", aggregated_by=AggregationType.WORD),
             SleepFrame(),  # Ensure messages are processed
             CancelFrame(),
         ]
@@ -401,8 +402,8 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
         frames_to_send = [
             BotStartedSpeakingFrame(),
             SleepFrame(),
-            TTSTextFrame(text="Assistant"),
+            TTSTextFrame(text="Assistant", aggregated_by=AggregationType.WORD),
-            TTSTextFrame(text="message"),
+            TTSTextFrame(text="message", aggregated_by=AggregationType.WORD),
             BotStoppedSpeakingFrame(),
         ]
 
@@ -439,7 +440,7 @@ class TestUserTranscriptProcessor(unittest.IsolatedAsyncioTestCase):
 
         # Test the specific pattern shared
         def make_tts_text_frame(text: str) -> TTSTextFrame:
-            frame = TTSTextFrame(text=text)
+            frame = TTSTextFrame(text=text, aggregated_by=AggregationType.WORD)
             frame.includes_inter_frame_spaces = True
             return frame
 

uv.lock

@@ -419,16 +419,30 @@ wheels = [
 
 [[package]]
 name = "aws-sdk-bedrock-runtime"
-version = "0.1.1"
+version = "0.2.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "smithy-aws-core", extra = ["eventstream", "json"], marker = "python_full_version >= '3.12'" },
     { name = "smithy-core", marker = "python_full_version >= '3.12'" },
     { name = "smithy-http", extra = ["awscrt"], marker = "python_full_version >= '3.12'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/1d/78/48574454b3cac869df67665e4a403ebfc3abfcfba2c2ff01ccfd67d55f8f/aws_sdk_bedrock_runtime-0.1.1.tar.gz", hash = "sha256:c896f99e675c3a1ab600633a07b785f3dc9fe8ab94f640b1f992b63da2dfc784", size = 82446, upload-time = "2025-10-21T20:25:25.845Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/db/94/f2451bb09c106e5690bbb88fc366637cdcec942b352ed9bb788804c877e0/aws_sdk_bedrock_runtime-0.2.0.tar.gz", hash = "sha256:8de52dd4492e74c73244d4b41a52304e1db368814a10e49dbbf8f4e8e412cd0e", size = 88156, upload-time = "2025-11-22T00:35:44.978Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/83/07/62c0b70223d178c138f29124ac2f7973a6ba803abc7735b6a01a85217f3d/aws_sdk_bedrock_runtime-0.1.1-py3-none-any.whl", hash = "sha256:c0336b377b2112cf88197d3d44302fbeb3efb1101989fa49ae55e78f49cfe345", size = 74954, upload-time = "2025-10-21T20:25:24.973Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/6b/07fbddd31dd6e38c967fe088b5e91a7cc3a2bc0f645f18b4e5d45bc03f1f/aws_sdk_bedrock_runtime-0.2.0-py3-none-any.whl", hash = "sha256:19594de50a52d199d73efca153c0a2328bd781827715a6e012d50b11085236cc", size = 79875, upload-time = "2025-11-22T00:35:44.092Z" },
+]
+
+[[package]]
+name = "aws-sdk-sagemaker-runtime-http2"
+version = "0.1.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "smithy-aws-core", extra = ["eventstream", "json"], marker = "python_full_version >= '3.12'" },
+    { name = "smithy-core", marker = "python_full_version >= '3.12'" },
+    { name = "smithy-http", extra = ["awscrt"], marker = "python_full_version >= '3.12'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/6e/ca/00f9c55887fc0f3fa345995dd871d40ff81473ab1591e56b4b4483d99d00/aws_sdk_sagemaker_runtime_http2-0.1.0.tar.gz", hash = "sha256:5077ec0c4440495b15004bbf04e27bc0bc137f1f8950d32195c6b45d7788d837", size = 20863, upload-time = "2025-11-22T00:20:56.358Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9c/24/2e2f727c51c20f4625cd19364d9421dbd7c893fe2b53a46eb0caaf6263a2/aws_sdk_sagemaker_runtime_http2-0.1.0-py3-none-any.whl", hash = "sha256:1aebb728ba6c6d14e58e29ecf89b51f7abbe8786d34144f8a7d59a419e80bd2f", size = 21911, upload-time = "2025-11-22T00:20:55.054Z" },
 ]
 
 [[package]]
@@ -4442,9 +4456,6 @@ assemblyai = [
 asyncai = [
     { name = "websockets" },
 ]
-auth = [
-    { name = "pyjwt" },
-]
 aws = [
     { name = "aioboto3" },
     { name = "websockets" },
@@ -4572,6 +4583,9 @@ runner = [
     { name = "python-dotenv" },
     { name = "uvicorn" },
 ]
+sagemaker = [
+    { name = "aws-sdk-sagemaker-runtime-http2", marker = "python_full_version >= '3.12'" },
+]
 sarvam = [
     { name = "sarvamai" },
     { name = "websockets" },
@@ -4657,7 +4671,8 @@ requires-dist = [
     { name = "aiortc", marker = "extra == 'webrtc'", specifier = ">=1.13.0,<2" },
     { name = "anthropic", marker = "extra == 'anthropic'", specifier = "~=0.49.0" },
     { name = "audioop-lts", marker = "python_full_version >= '3.13'", specifier = "~=0.2.1" },
-    { name = "aws-sdk-bedrock-runtime", marker = "python_full_version >= '3.12' and extra == 'aws-nova-sonic'", specifier = "~=0.1.1" },
+    { name = "aws-sdk-bedrock-runtime", marker = "python_full_version >= '3.12' and extra == 'aws-nova-sonic'", specifier = "~=0.2.0" },
+    { name = "aws-sdk-sagemaker-runtime-http2", marker = "python_full_version >= '3.12' and extra == 'sagemaker'" },
     { name = "azure-cognitiveservices-speech", marker = "extra == 'azure'", specifier = "~=1.42.0" },
     { name = "cartesia", marker = "extra == 'cartesia'", specifier = "~=2.0.3" },
     { name = "coremltools", marker = "extra == 'local-smart-turn'", specifier = ">=8.0" },
@@ -4724,7 +4739,6 @@ requires-dist = [
     { name = "pyaudio", marker = "extra == 'local'", specifier = "~=0.2.14" },
     { name = "pydantic", specifier = ">=2.10.6,<3" },
     { name = "pygobject", marker = "extra == 'gstreamer'", specifier = "~=3.50.0" },
-    { name = "pyjwt", marker = "extra == 'auth'", specifier = ">=2.10.1" },
     { name = "pyloudnorm", specifier = "~=0.1.1" },
     { name = "python-dotenv", marker = "extra == 'runner'", specifier = ">=1.0.0,<2.0.0" },
     { name = "pyvips", extras = ["binary"], marker = "extra == 'moondream'", specifier = "~=3.0.0" },
@@ -4749,7 +4763,7 @@ requires-dist = [
     { name = "wait-for2", marker = "python_full_version < '3.12'", specifier = ">=0.4.1" },
     { name = "websockets", marker = "extra == 'websockets-base'", specifier = ">=13.1,<16.0" },
 ]
-provides-extras = ["aic", "anthropic", "assemblyai", "asyncai", "aws", "aws-nova-sonic", "azure", "cartesia", "cerebras", "deepseek", "daily", "deepgram", "elevenlabs", "fal", "fireworks", "fish", "gladia", "google", "grok", "groq", "gstreamer", "heygen", "hume", "inworld", "krisp", "koala", "langchain", "livekit", "lmnt", "local", "mcp", "mem0", "mistral", "mlx-whisper", "moondream", "nim", "neuphonic", "noisereduce", "openai", "openpipe", "openrouter", "perplexity", "playht", "qwen", "rime", "riva", "runner", "sambanova", "sarvam", "sentry", "local-smart-turn", "local-smart-turn-v3", "remote-smart-turn", "silero", "simli", "soniox", "soundfile", "speechmatics", "strands", "tavus", "together", "tracing", "ultravox", "webrtc", "websocket", "websockets-base", "whisper", "auth"]
+provides-extras = ["aic", "anthropic", "assemblyai", "asyncai", "aws", "aws-nova-sonic", "azure", "cartesia", "cerebras", "deepseek", "daily", "deepgram", "elevenlabs", "fal", "fireworks", "fish", "gladia", "google", "grok", "groq", "gstreamer", "heygen", "hume", "inworld", "krisp", "koala", "langchain", "livekit", "lmnt", "local", "mcp", "mem0", "mistral", "mlx-whisper", "moondream", "nim", "neuphonic", "noisereduce", "openai", "openpipe", "openrouter", "perplexity", "playht", "qwen", "rime", "riva", "runner", "sambanova", "sarvam", "sentry", "local-smart-turn", "local-smart-turn-v3", "remote-smart-turn", "sagemaker", "silero", "simli", "soniox", "soundfile", "speechmatics", "strands", "tavus", "together", "tracing", "ultravox", "webrtc", "websocket", "websockets-base", "whisper"]
 
 [package.metadata.requires-dev]
 dev = [
@@ -6526,16 +6540,16 @@ wheels = [
 
 [[package]]
 name = "smithy-aws-core"
-version = "0.1.1"
+version = "0.2.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "aws-sdk-signers", marker = "python_full_version >= '3.12'" },
     { name = "smithy-core", marker = "python_full_version >= '3.12'" },
     { name = "smithy-http", marker = "python_full_version >= '3.12'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/56/d3/f847e0fd36b95aa36ce3a4c9ce1a08e16b2aa9a56b71714045c9c924e282/smithy_aws_core-0.1.1.tar.gz", hash = "sha256:78dfd7040fc2bc72b6af293096642fc9a7bfd2db28ddbdf7c4110535eab9d662", size = 11196, upload-time = "2025-10-21T20:21:18.648Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/c1/c8/5970c869527972b23a1733de3993d54283c84a2340e84acdd48a11aa0ff4/smithy_aws_core-0.2.0.tar.gz", hash = "sha256:dfa1ecd311d6f0a16f48c86d793085e2a0a33a46de897d129dd1f142a43bf7f6", size = 11344, upload-time = "2025-11-21T18:33:01.928Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d0/04/87cb06f0f6d664b5cffdef6d4042dd52c11c138436084d30ffdaa3543031/smithy_aws_core-0.1.1-py3-none-any.whl", hash = "sha256:0d1634f276c2999dc2a04fafef63b9d28309de50d939d1d49df952773a7063c4", size = 18963, upload-time = "2025-10-21T20:21:17.692Z" },
+    { url = "https://files.pythonhosted.org/packages/88/25/739c0005a6cb4effbc2d37fe23590660b508fe314200f4acf94410a8f315/smithy_aws_core-0.2.0-py3-none-any.whl", hash = "sha256:d112082ef77758e1977f8694cf690ac35c76570c12a6590fccd5da085a3ce507", size = 18966, upload-time = "2025-11-21T18:33:00.812Z" },
 ]
 
 [package.optional-dependencies]
@@ -6548,35 +6562,35 @@ json = [
 
 [[package]]
 name = "smithy-aws-event-stream"
-version = "0.1.0"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "smithy-core", marker = "python_full_version >= '3.12'" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/49/26/8ff24194efed60b2df18f610ea05fa2a4c6546858b80a0a51335a4943b9b/smithy_aws_event_stream-0.1.0.tar.gz", hash = "sha256:6634691a3bf5d4801a2c29f0761db2dc4771f3ae43cdee50c10d4b4bb2f86475", size = 12216, upload-time = "2025-09-29T19:37:14.659Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/90/c4/2b63d31af58fc359577e5515bf730348a235f2f2fa10e17af8640495c81c/smithy_aws_event_stream-0.1.0-py3-none-any.whl", hash = "sha256:17a7300a85cb90df4c6c23f895ca6343361fa419203c3cf80019edd7d3b5f036", size = 15581, upload-time = "2025-09-29T19:37:13.589Z" },
-]
-
-[[package]]
-name = "smithy-core"
-version = "0.1.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/b9/8d/16028d03456071d21de7591f1e1e6a1cc81b2389e53ef8663dbf59caf9cd/smithy_core-0.1.0.tar.gz", hash = "sha256:b159b8905264e1e4c613eab9f74cec0b2f5b8119c42fbadddb4da0a8ed8050e9", size = 48415, upload-time = "2025-09-29T19:37:16.873Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/ca/5b/563cb2beadcfa40597b0c3ff3f2d42e21f065b14782c4ba9cb41a44b745f/smithy_core-0.1.0-py3-none-any.whl", hash = "sha256:cb44e9355fb89e89f2c6ba6a1d59c5db4f2f7282c72d31d9307b6202d66cd0fa", size = 62895, upload-time = "2025-09-29T19:37:15.917Z" },
-]
-
-[[package]]
-name = "smithy-http"
 version = "0.2.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "smithy-core", marker = "python_full_version >= '3.12'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/3c/1c/44e99a7dfb8c39bf0c3d998accdf4573a7a3488863b90f21af260cec2d45/smithy_http-0.2.0.tar.gz", hash = "sha256:2382562fa9af326be455f14b18615f16ffe9db756e51b2a4ca0d23e1b881cff8", size = 26729, upload-time = "2025-10-21T20:21:06.146Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/31/90/78283c21484f8cf9862982e53bc2769b784910735fb5fb2400a17bfb5fdd/smithy_aws_event_stream-0.2.0.tar.gz", hash = "sha256:99700a11346e7ab1435ff2e53e6f6d60a1e857f2b2ee1941d40b54270adf3323", size = 12278, upload-time = "2025-11-21T18:33:03.79Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d4/e2/d475fad81ac74ec0e145cb6d72afe5ecde4e2358bd632c2fd5d3f4bc87dc/smithy_http-0.2.0-py3-none-any.whl", hash = "sha256:49ee2402d7737798d70f99f491fbfb2a5767283ae562e21b6f86e3fd14f3e3e0", size = 37328, upload-time = "2025-10-21T20:21:05.362Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/f5/08b997eee81b55150496ce565f0e03c72d0c80e5b218170bdeae7c46a5a4/smithy_aws_event_stream-0.2.0-py3-none-any.whl", hash = "sha256:679a0c7d944e67d3a55d287541b3ca1e61f9d6a62e13401367dcc034e75aa55d", size = 15567, upload-time = "2025-11-21T18:33:02.711Z" },
+]
+
+[[package]]
+name = "smithy-core"
+version = "0.2.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/c7/f6/140f0be9331dd7cd8fa012b3ca4735df39a1a81d03eea89728f997249116/smithy_core-0.2.0.tar.gz", hash = "sha256:05c3e3309df5dcb9cf53e241bd57a96510e4575186443ea157db9dbb59b6c85e", size = 50334, upload-time = "2025-11-21T18:33:05.697Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/16/e3/d0defa2acf50b91625fe15e3ddb0c8e41ff64363a1f4cd9b8f19ae2ec0c6/smithy_core-0.2.0-py3-none-any.whl", hash = "sha256:db4620da3497abb60f79ac1d8a738d3eac46d7e820bfb50c777c36e932915239", size = 64777, upload-time = "2025-11-21T18:33:04.591Z" },
+]
+
+[[package]]
+name = "smithy-http"
+version = "0.3.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "smithy-core", marker = "python_full_version >= '3.12'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/1c/c7/4d8be56e897f99f3b6ffcdf52ba00a468febc939fca85b90f1c122450830/smithy_http-0.3.0.tar.gz", hash = "sha256:55dcc3af315eee6863d2f3f58ada1d9cb4bcc3a57faac10a1b21d4a93722f520", size = 28674, upload-time = "2025-11-21T18:33:07.387Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/2d/e5/59ae79ecdc9a935ad10512c581b3054ebb1afd90498ecc8afaf141dbc22b/smithy_http-0.3.0-py3-none-any.whl", hash = "sha256:972924304febd77c7134a7cffab83ce3b48423ff966dcc1f257e2c0d58fa9b18", size = 40520, upload-time = "2025-11-21T18:33:06.312Z" },
 ]
 
 [package.optional-dependencies]
@@ -6586,15 +6600,15 @@ awscrt = [
 
 [[package]]
 name = "smithy-json"
-version = "0.1.0"
+version = "0.2.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "ijson", marker = "python_full_version >= '3.12'" },
     { name = "smithy-core", marker = "python_full_version >= '3.12'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/e2/5b/0ecb10007475e1b8faca3bbff1be2fc6edb3ea12ffc5e939e2249be95325/smithy_json-0.1.0.tar.gz", hash = "sha256:84fb48e445b87d850c240d837702c16b259ea53bad76b655ac1bbd8094d48912", size = 7086, upload-time = "2025-09-29T19:37:20.432Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/89/cf/e319a2a299b27bc0addf46ee3d4b9c25ec0817e3a0507b2b7a33eddc19f1/smithy_json-0.2.0.tar.gz", hash = "sha256:0946066fdda15d6a579dfdd4b61a547ab915eb057bd176fc2bc17d01dc789499", size = 7157, upload-time = "2025-11-21T18:33:08.968Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/62/95/e11c04e56aae12b62e38c49000004a1dc598a64dc207018c08448efde322/smithy_json-0.1.0-py3-none-any.whl", hash = "sha256:80ff64734dccdabf1ba6a2908555b97e60f62c07c3a27df48e421ee058413cb9", size = 9914, upload-time = "2025-09-29T19:37:19.459Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/b1/33012ac5b2e5940a00b6e1ccc313330e6f8692152a151f72a398cd6be0e0/smithy_json-0.2.0-py3-none-any.whl", hash = "sha256:5018a4e61731afa3094a02d737d4f956dbf270c271410c089045a17d86fc3b3b", size = 9911, upload-time = "2025-11-21T18:33:08.267Z" },
 ]
 
 [[package]]