wx44wx/pipecat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

changelog: add PR prefix to PR link

Browse Source
This commit is contained in:
Aleix Conchillo Flaqué
2025-12-17 11:15:28 -08:00
committed by Mark Backman
parent 491d298c10
commit 92970c7873
2 changed files with 81 additions and 73 deletions
Download Patch File Download Diff File Expand all files Collapse all files

150
CHANGELOG.md
View File

@@ -18,37 +18,37 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
- Added additional functionality related to "thinking", for Google and - Added additional functionality related to "thinking", for Google and
Anthropic LLMs. Anthropic LLMs.
1. New typed parameters for Google and Anthropic LLMs that control the 1. New typed parameters for Google and Anthropic LLMs that control the
models' thinking behavior (like how much thinking to do, and whether to models' thinking behavior (like how much thinking to do, and whether to
output thoughts or thought summaries): output thoughts or thought summaries):
- `AnthropicLLMService.ThinkingConfig` - `AnthropicLLMService.ThinkingConfig`
- `GoogleLLMService.ThinkingConfig` - `GoogleLLMService.ThinkingConfig`
2. New frames for representing thoughts output by LLMs: 2. New frames for representing thoughts output by LLMs:
- `LLMThoughtStartFrame` - `LLMThoughtStartFrame`
- `LLMThoughtTextFrame` - `LLMThoughtTextFrame`
- `LLMThoughtEndFrame` - `LLMThoughtEndFrame`
3. A generic mechanism for recording LLM thoughts to context, used 3. A generic mechanism for recording LLM thoughts to context, used
specifically to support Anthropic, whose thought signatures are expected specifically to support Anthropic, whose thought signatures are expected
to appear alongside the text of the thoughts within assistant context to appear alongside the text of the thoughts within assistant context
messages. See: messages. See:
- `LLMThoughtEndFrame.signature` - `LLMThoughtEndFrame.signature`
- `LLMAssistantAggregator` handling of the above field - `LLMAssistantAggregator` handling of the above field
- `AnthropicLLMAdapter` handling of `"thought"` context messages - `AnthropicLLMAdapter` handling of `"thought"` context messages
4. Google-specific logic for inserting thought signatures into the context, 4. Google-specific logic for inserting thought signatures into the context,
to help maintain thinking continuity in a chain of LLM calls. See: to help maintain thinking continuity in a chain of LLM calls. See:
- `GoogleLLMService` sending `LLMMessagesAppendFrame`s to add - `GoogleLLMService` sending `LLMMessagesAppendFrame`s to add
LLM-specific LLM-specific
`"thought_signature"` messages to context `"thought_signature"` messages to context
- `GeminiLLMAdapter` handling of `"thought_signature"` messages - `GeminiLLMAdapter` handling of `"thought_signature"` messages
5. An expansion of `TranscriptProcessor` to process LLM thoughts in 5. An expansion of `TranscriptProcessor` to process LLM thoughts in
addition to user and assistant utterances. See: addition to user and assistant utterances. See:
- `TranscriptProcessor(process_thoughts=True)` (defaults to `False`) - `TranscriptProcessor(process_thoughts=True)` (defaults to `False`)
- `ThoughtTranscriptionMessage`, which is now also emitted with the - `ThoughtTranscriptionMessage`, which is now also emitted with the
`"on_transcript_update"` event `"on_transcript_update"` event
(PR [#3175](https://github.com/pipecat-ai/pipecat/pull/3175)) (PR [#3175](https://github.com/pipecat-ai/pipecat/pull/3175))
- Data and control frames can now be marked as non-interruptible by using the - Data and control frames can now be marked as non-interruptible by using the
`UninterruptibleFrame` mixin. Frames marked as `UninterruptibleFrame` will `UninterruptibleFrame` mixin. Frames marked as `UninterruptibleFrame` will
not be interrupted during processing, and any queued frames of this type will not be interrupted during processing, and any queued frames of this type will
be retained in the internal queues. This is useful when you need ordered be retained in the internal queues. This is useful when you need ordered
frames (data or control) that should not be discarded or cancelled due to frames (data or control) that should not be discarded or cancelled due to
@@ -67,28 +67,30 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
- Added to `AWSNovaSonicLLMService` functionality related to the new (and now - Added to `AWSNovaSonicLLMService` functionality related to the new (and now
default) Nova 2 Sonic model (`"amazon.nova-2-sonic-v1:0"`): default) Nova 2 Sonic model (`"amazon.nova-2-sonic-v1:0"`):
- Added the `endpointing_sensitivity` parameter to control how quickly the
model decides the user has stopped speaking. - Added the `endpointing_sensitivity` parameter to control how quickly the
- Made the assistant-response-trigger hack a no-op. It's only needed for model decides the user has stopped speaking.
the older Nova Sonic model. - Made the assistant-response-trigger hack a no-op. It's only needed for
(PR [#3212](https://github.com/pipecat-ai/pipecat/pull/3212)) the older Nova Sonic model.
(PR [#3212](https://github.com/pipecat-ai/pipecat/pull/3212))
- [Ultravox Realtime](https://docs.ultravox.ai) is now a supported - [Ultravox Realtime](https://docs.ultravox.ai) is now a supported
speech-to-speech service. speech-to-speech service.
- Added `UltravoxRealtimeLLMService` for the integration.
- Added `49-ultravox-realtime.py` example (with tool calling). - Added `UltravoxRealtimeLLMService` for the integration.
(PR [#3227](https://github.com/pipecat-ai/pipecat/pull/3227)) - Added `49-ultravox-realtime.py` example (with tool calling).
(PR [#3227](https://github.com/pipecat-ai/pipecat/pull/3227))
- Added Daily PSTN dial-in support to the development runner with `--dialin` - Added Daily PSTN dial-in support to the development runner with `--dialin`
flag. This includes: flag. This includes:
- `/daily-dialin-webhook` endpoint that handles incoming Daily PSTN webhooks - `/daily-dialin-webhook` endpoint that handles incoming Daily PSTN webhooks
- Automatic Daily room creation with SIP configuration - Automatic Daily room creation with SIP configuration
- `DialinSettings` and `DailyDialinRequest` types in `pipecat.runner.types` - `DialinSettings` and `DailyDialinRequest` types in `pipecat.runner.types`
for type-safe dial-in data for type-safe dial-in data
- The runner now mimics Pipecat Cloud's dial-in webhook handling for local - The runner now mimics Pipecat Cloud's dial-in webhook handling for local
development development
(PR [#3235](https://github.com/pipecat-ai/pipecat/pull/3235)) (PR [#3235](https://github.com/pipecat-ai/pipecat/pull/3235))
- Add Gladia session id to logs for `GladiaSTTService`. - Add Gladia session id to logs for `GladiaSTTService`.
(PR [#3236](https://github.com/pipecat-ai/pipecat/pull/3236)) (PR [#3236](https://github.com/pipecat-ai/pipecat/pull/3236))
@@ -128,13 +130,15 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
APIs (Interactive Avatar and Live Avatar). APIs (Interactive Avatar and Live Avatar).
Using them is as simple as specifying the `service_type` when creating the Using them is as simple as specifying the `service_type` when creating the
`HeyGenVideoService` and the `HeyGenTransport`: `HeyGenVideoService` and the `HeyGenTransport`:
```python
heyGen = HeyGenVideoService( ```python
api_key=os.getenv("HEYGEN_LIVE_AVATAR_API_KEY"), heyGen = HeyGenVideoService(
service_type=ServiceType.LIVE_AVATAR, api_key=os.getenv("HEYGEN_LIVE_AVATAR_API_KEY"),
session=session, service_type=ServiceType.LIVE_AVATAR,
) session=session,
``` )
```
(PR [#3210](https://github.com/pipecat-ai/pipecat/pull/3210)) (PR [#3210](https://github.com/pipecat-ai/pipecat/pull/3210))
- Made `"amazon.nova-2-sonic-v1:0"` the new default model for - Made `"amazon.nova-2-sonic-v1:0"` the new default model for
@@ -148,9 +152,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
(PR [#3214](https://github.com/pipecat-ai/pipecat/pull/3214)) (PR [#3214](https://github.com/pipecat-ai/pipecat/pull/3214))
- Updated default models for: - Updated default models for:
- `GeminiLiveLLMService` to `gemini-2.5-flash-native-audio-preview-12-2025`.
- `GeminiLiveVertexLLMService` to `gemini-live-2.5-flash-native-audio`. - `GeminiLiveLLMService` to `gemini-2.5-flash-native-audio-preview-12-2025`.
(PR [#3228](https://github.com/pipecat-ai/pipecat/pull/3228)) - `GeminiLiveVertexLLMService` to `gemini-live-2.5-flash-native-audio`.
(PR [#3228](https://github.com/pipecat-ai/pipecat/pull/3228))
- Changed the `reason` field in `EndFrame`, `CancelFrame`, `EndTaskFrame`, and - Changed the `reason` field in `EndFrame`, `CancelFrame`, `EndTaskFrame`, and
`CancelTaskFrame` from `str` to `Any` to indicate that it can hold values `CancelTaskFrame` from `str` to `Any` to indicate that it can hold values
@@ -160,33 +165,36 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
- Updated websocket STT services to use the `WebsocketSTTService` base class. - Updated websocket STT services to use the `WebsocketSTTService` base class.
This base class manages the websocket connection and handles reconnects. This base class manages the websocket connection and handles reconnects.
Updated services: Updated services:
- `AssemblyAISTTService`
- `AWSTranscribeSTTService` - `AssemblyAISTTService`
- `GladiaSTTService` - `AWSTranscribeSTTService`
- `SonioxSTTService` - `GladiaSTTService`
(PR [#3236](https://github.com/pipecat-ai/pipecat/pull/3236)) - `SonioxSTTService`
(PR [#3236](https://github.com/pipecat-ai/pipecat/pull/3236))
- Changed Inworld's TTS service implementations: - Changed Inworld's TTS service implementations:
- Previously, the HTTP implementation was named `InworldTTSService`. That
has been moved to `InworldHttpTTSService`. This service now supports - Previously, the HTTP implementation was named `InworldTTSService`. That
word-timestamp alignment data in both streaming and non-streaming modes. has been moved to `InworldHttpTTSService`. This service now supports
- Updated the `InworldTTSService` class to use Inworld's Websocket API. word-timestamp alignment data in both streaming and non-streaming modes.
This class now has support for word-timestamp alignment data and tracks - Updated the `InworldTTSService` class to use Inworld's Websocket API.
contexts for each user turn. This class now has support for word-timestamp alignment data and tracks
(PR [#3239](https://github.com/pipecat-ai/pipecat/pull/3239)) contexts for each user turn.
(PR [#3239](https://github.com/pipecat-ai/pipecat/pull/3239))
- ⚠️ Breaking change: `WordTTSService.start_word_timestamps()` and - ⚠️ Breaking change: `WordTTSService.start_word_timestamps()` and
`WordTTSService.reset_word_timestamps()` are now async. `WordTTSService.reset_word_timestamps()` are now async.
(PR [#3240](https://github.com/pipecat-ai/pipecat/pull/3240)) (PR [#3240](https://github.com/pipecat-ai/pipecat/pull/3240))
- Updated the current RTVI version to 1.1.0 to reflect recent additions and - Updated the current RTVI version to 1.1.0 to reflect recent additions and
deprecations. deprecations.
- New RTVI Messages: `send-text` and `bot-output`
- Deprecated Messages: `append-to-context` and `bot-transcription` - New RTVI Messages: `send-text` and `bot-output`
(PR [#3248](https://github.com/pipecat-ai/pipecat/pull/3248)) - Deprecated Messages: `append-to-context` and `bot-transcription`
(PR [#3248](https://github.com/pipecat-ai/pipecat/pull/3248))
- `MoondreamService` now pushes `VisionFullResponseStartFrame`, - `MoondreamService` now pushes `VisionFullResponseStartFrame`,
`VisionFullResponseEndFrame` and `VisionTextFrame`. `VisionFullResponseEndFrame` and `VisionTextFrame`.
(PR [#3252](https://github.com/pipecat-ai/pipecat/pull/3252)) (PR [#3252](https://github.com/pipecat-ai/pipecat/pull/3252))
### Deprecated ### Deprecated

4
changelog/_template.md.j2
View File

@@ -4,7 +4,8 @@
### {{ definitions[category]['name'] }} ### {{ definitions[category]['name'] }}
{% for text, values in sections[section][category].items() %} {% for text, values in sections[section][category].items() %}
{{ text }} ({{ values|join(', ') }}) {{ text }}
(PR {{ values|join(', ') }})
{% endfor %} {% endfor %}
{% endfor %} {% endfor %}
@@ -13,4 +14,3 @@ No significant changes.
{% endif %} {% endif %}
{% endfor %} {% endfor %}