feat(voice_agent): pipecat 0.0.103 → 0.0.106 - #248
Merged
vizsatiz merged 3 commits intoJun 22, 2026
Conversation
- Bump pipecat-ai to 0.0.106, remove dropped `local-smart-turn-v3` extra - Replace deprecated InputParams with typed Settings classes across tts_service and stt_service - Replace ParallelPipeline-based multi-language routing (LanguageSwitcher/STTLanguageSwitcher) with single TTS/STT service + runtime TTSUpdateSettingsFrame/STTUpdateSettingsFrame - Fix spurious double tool call on language switch: mutate context system message in-place instead of using LLMMessagesUpdateFrame (which stripped in-flight tool call from history) - Remove deprecated TranscriptProcessor from pipeline (redundant with existing aggregator events)
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: ⛔ Files ignored due to path filters (1)
📒 Files selected for processing (1)
✅ Files skipped from review due to trivial changes (1)
📝 WalkthroughWalkthroughRefactors multi-language switching architecture: removes per-language switcher pipelines and replaces them with single TTS/STT service instances per call, each configured with provider-specific settings objects instead of option/parameter construction. Introduces conditional queuing of TTS/STT language-update frames at runtime via new factory methods. Modifies conversation context mutation to update system messages in-place instead of snapshot-replacing via queued frames. Updates pipecat-ai dependency to 0.0.106. ChangesMulti-Language Switching Refactoring
Sequence DiagramsequenceDiagram
autonumber
participant Client
participant PipecatService as Pipecat Service
participant LangDetectionTool as Language Detection Tool
participant TTSFactory as TTS Factory
participant STTFactory as STT Factory
participant FrameQueue as Task Queue
Client->>PipecatService: Start session (default_language)
PipecatService->>PipecatService: Create single TTS & STT services (settings-based)
PipecatService->>LangDetectionTool: Configure with tts_provider, stt_provider, tts_voice_ids
Note over Client,FrameQueue: User audio received
Client->>PipecatService: Send user audio
PipecatService->>LangDetectionTool: Detect language
LangDetectionTool->>TTSFactory: create_language_update_frame(tts_provider, lang, voice_id)
TTSFactory-->>LangDetectionTool: TTSUpdateSettingsFrame or None
LangDetectionTool->>STTFactory: create_language_update_frame(stt_provider, lang)
STTFactory-->>LangDetectionTool: STTUpdateSettingsFrame or None
LangDetectionTool->>FrameQueue: Queue non-null update frames
LangDetectionTool->>LangDetectionTool: Mutate context.messages in-place (system role)
FrameQueue->>PipecatService: Apply queued settings updates to services
PipecatService->>Client: Respond (using updated language/voice)
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Possibly related PRs
Suggested reviewers
Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
wavefront/server/apps/call_processing/call_processing/services/pipecat_service.py (1)
23-23:⚠️ Potential issue | 🔴 CriticalAdd missing
local-smart-turn-v3extra to pipecat-ai dependency.The
local-smart-turn-v3extra was removed from the pyproject.toml pipecat-ai dependency (currently:pipecat-ai[websocket,cartesia,google,silero,deepgram,groq,runner,azure,sarvam,tracing]==0.0.106), but the code imports and usesLocalSmartTurnAnalyzerV3on line 23 and line 412. This import requires the extra to be present; without it, the application will fail at runtime. Either restore the extra in the dependency or remove this usage from the code.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@wavefront/server/apps/call_processing/call_processing/services/pipecat_service.py` at line 23, The project imports and uses LocalSmartTurnAnalyzerV3 (from pipecat.audio.turn.smart_turn.local_smart_turn_v3) but the pipecat-ai dependency no longer includes the local-smart-turn-v3 extra; restore the extra in the pyproject.toml dependency declaration (add local-smart-turn-v3 to pipecat-ai[...] version 0.0.106) so the import in pipecat_service.py and its usage of LocalSmartTurnAnalyzerV3 work at runtime, or alternatively remove/replace all uses of LocalSmartTurnAnalyzerV3 in the codebase if you intend to drop that extra.
🧹 Nitpick comments (2)
wavefront/server/apps/call_processing/call_processing/services/tts_service.py (1)
274-277: PEP 484: Use explicitNonetype annotation.The
voice_idparameter uses implicitOptionalwhich is prohibited by PEP 484 and flagged by Ruff (RUF013).♻️ Suggested fix
`@staticmethod` def create_language_update_frame( - provider: str, lang_code: str, voice_id: str = None + provider: str, lang_code: str, voice_id: str | None = None ):🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@wavefront/server/apps/call_processing/call_processing/services/tts_service.py` around lines 274 - 277, The function create_language_update_frame has an implicitly optional parameter voice_id which violates PEP 484/RUF013; update its signature to annotate voice_id explicitly as Optional[str] (i.e., voice_id: Optional[str] = None) and add or update the typing import (from typing import Optional) at the top of the module if missing so the type is resolvable.wavefront/server/apps/call_processing/call_processing/services/language_detection_tool.py (1)
174-176: Assumption thatmessages[0]is always the system message may be fragile.The code directly assigns to
current_messages[0]assuming the system message is at index 0. While this is likely true in practice given how messages are initialized inpipecat_service.py, consider adding a defensive check or comment documenting this assumption.♻️ Suggested defensive approach
current_messages = context.get_messages() - current_messages[0] = {'role': 'system', 'content': updated_content} + # System message is always at index 0 (initialized in pipecat_service.py) + if current_messages and current_messages[0].get('role') == 'system': + current_messages[0] = {'role': 'system', 'content': updated_content} + else: + logger.warning('Expected system message at index 0, skipping prompt update') context.set_messages(current_messages)🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@wavefront/server/apps/call_processing/call_processing/services/language_detection_tool.py` around lines 174 - 176, The current code assumes the system message is always at index 0 by directly writing to current_messages[0]; instead, retrieve messages via context.get_messages(), check that a message with 'role' == 'system' exists and update that entry (or if none exists, insert/prepend a new system message with the updated_content), then call context.set_messages(current_messages); use the functions/variables current_messages, context.get_messages, context.set_messages and do not blindly overwrite messages[0].
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In
`@wavefront/server/apps/call_processing/call_processing/services/language_detection_tool.py`:
- Around line 129-147: The log is misleading because
create_language_update_frame can return None; update the block around
TTSServiceFactory.create_language_update_frame and
STTServiceFactory.create_language_update_frame to record whether each frame was
actually created/queued (e.g., tts_frame_queued, stt_frame_queued booleans),
call task.queue_frame only when frame exists, and change logger.info to reflect
the real outcome (both queued, only TTS queued, only STT queued, or none queued)
and add logger.warning/error when neither frame was queued; reference the
TTSServiceFactory.create_language_update_frame,
STTServiceFactory.create_language_update_frame, task.queue_frame, logger.info,
and tts_voice_ids symbols to locate and update the logic.
---
Outside diff comments:
In
`@wavefront/server/apps/call_processing/call_processing/services/pipecat_service.py`:
- Line 23: The project imports and uses LocalSmartTurnAnalyzerV3 (from
pipecat.audio.turn.smart_turn.local_smart_turn_v3) but the pipecat-ai dependency
no longer includes the local-smart-turn-v3 extra; restore the extra in the
pyproject.toml dependency declaration (add local-smart-turn-v3 to
pipecat-ai[...] version 0.0.106) so the import in pipecat_service.py and its
usage of LocalSmartTurnAnalyzerV3 work at runtime, or alternatively
remove/replace all uses of LocalSmartTurnAnalyzerV3 in the codebase if you
intend to drop that extra.
---
Nitpick comments:
In
`@wavefront/server/apps/call_processing/call_processing/services/language_detection_tool.py`:
- Around line 174-176: The current code assumes the system message is always at
index 0 by directly writing to current_messages[0]; instead, retrieve messages
via context.get_messages(), check that a message with 'role' == 'system' exists
and update that entry (or if none exists, insert/prepend a new system message
with the updated_content), then call context.set_messages(current_messages); use
the functions/variables current_messages, context.get_messages,
context.set_messages and do not blindly overwrite messages[0].
In
`@wavefront/server/apps/call_processing/call_processing/services/tts_service.py`:
- Around line 274-277: The function create_language_update_frame has an
implicitly optional parameter voice_id which violates PEP 484/RUF013; update its
signature to annotate voice_id explicitly as Optional[str] (i.e., voice_id:
Optional[str] = None) and add or update the typing import (from typing import
Optional) at the top of the module if missing so the type is resolvable.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: f9cdf43d-cb52-45c0-a40c-2da88a28c3f3
⛔ Files ignored due to path filters (1)
wavefront/server/uv.lockis excluded by!**/*.lock
📒 Files selected for processing (5)
wavefront/server/apps/call_processing/call_processing/services/language_detection_tool.pywavefront/server/apps/call_processing/call_processing/services/pipecat_service.pywavefront/server/apps/call_processing/call_processing/services/stt_service.pywavefront/server/apps/call_processing/call_processing/services/tts_service.pywavefront/server/apps/call_processing/pyproject.toml
vizsatiz
approved these changes
Mar 24, 2026
vizsatiz
previously approved these changes
Jun 21, 2026
…o-00106 # Conflicts: # wavefront/server/uv.lock
vizsatiz
approved these changes
Jun 22, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
local-smart-turn-v3extraSummary by CodeRabbit