wx44wx/pipecat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
9,290 Commits 276 Branches 111 Tags
fe3af5d9f7edcb65f646d845e596ee467a1fb812
Commit Graph

9290 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Mark Backman
2ce9179662 Merge pull request #3958 from pipecat-ai/mb/deepgram-tts-audio-context 
Route Deepgram WebSocket TTS audio through audio context queue
 2026-03-10 11:37:39 -04:00
Mark Backman
50cc01a578 Guard against None context ID in append_to_audio_context 
After interruption, both _playing_context_id and _turn_context_id are
None. If a subclass calls append_to_audio_context(None, frame), the
recovery path matches (None == None) and creates a bogus audio context
that blocks the handler from ever processing the real context.

Early-return when context_id is falsy to prevent this.
 2026-03-10 11:34:03 -04:00
Mark Backman
d5c0789ab5 Add changelog for #3958 2026-03-10 11:34:03 -04:00
Mark Backman
92b5185165 Route Deepgram WebSocket TTS audio through audio context queue 
The Deepgram TTS service was bypassing pipecats audio context management
system, pushing audio frames directly via push_frame() instead of routing
them through append_to_audio_context(). This caused stale audio to leak
into the pipeline after interruptions and missed ordered playback
guarantees.

- Route audio frames through append_to_audio_context() with context
  availability checks to discard stale post-interruption frames
- Handle Flushed responses by appending TTSStoppedFrame and removing
  the audio context to signal completion
- Replace _handle_interruption override with on_audio_context_interrupted
  hook (the recommended pattern used by ElevenLabs and Cartesia)
- Remove redundant process_frame override that caused double-flush
  (base class already flushes via on_turn_context_completed)
- Remove redundant start_tts_usage_metrics call (base class handles
  aggregated usage metrics)
 2026-03-10 11:34:03 -04:00
sysradium
ba0ebd5525 Treat conversation_already_has_active_response as non-fatal in Realtime API 2026-03-10 15:57:55 +01:00
Gökmen Görgen
3a6f848a5b update test description. 2026-03-10 14:54:49 +01:00
kompfner
c660152a84 Merge pull request #3966 from pipecat-ai/pk/add-some-more-missing-55-examples 
Add missing 55-* update-settings examples for OpenPipe LLM and XTTS TTS
 2026-03-10 09:45:58 -04:00
Gökmen Görgen
a96702acfc fix test. 2026-03-10 14:41:18 +01:00
Gökmen Görgen
780559dc32 address feedback. 2026-03-10 14:23:00 +01:00
Gökmen Görgen
8e1c8a38e4 don't change enhancement level if bypass toggled. 2026-03-10 14:18:45 +01:00
Gökmen Görgen
483f6689ed address feedback, use one logging. 2026-03-10 13:52:13 +01:00
Gökmen Görgen
bc11bf9673 remove _is_filter_enabled from AICFilter and refactor related logic and tests. 2026-03-10 13:48:32 +01:00
Gökmen Görgen
82b300298a add changelog. 2026-03-10 13:36:15 +01:00
Gökmen Görgen
0c87fcc48c re-add bypass parameter support to AICFilter and update related unit tests. 2026-03-10 13:36:15 +01:00
Gökmen Görgen
df64f3f943 add enhancement_level support to AICFilter. 
# Conflicts:
#	src/pipecat/audio/filters/aic_filter.py
 2026-03-10 13:36:15 +01:00
Mark Backman
db22bf0f75 Merge pull request #3973 from yuki901/fish-audio-s2-pro 
Update Fish Audio default model from s1 to s2-pro
 2026-03-10 07:57:27 -04:00
Mark Backman
edc65fc45e Add changelog for #3974 2026-03-10 07:48:02 -04:00
Mark Backman
233867fdfb Make region optional and validate Azure STT config 
Make `region` optional so users can provide only `private_endpoint`.
Raise ValueError if neither is provided, and warn if both are given
(private_endpoint takes priority).
 2026-03-10 07:47:05 -04:00
yukiobata1
ceb53e044b Add changelog for #3973 2026-03-10 19:29:47 +09:00
yukiobata1
c7ef23dd22 Update Fish Audio default model from s1 to s2-pro 2026-03-10 18:22:20 +09:00
Kinshuk Bairagi
9cc2644719 Improve system message extraction in traced_llm 
Enhanced the logic for extracting the system message in the traced_llm decorator to support LLMContext via adapter and handle exceptions gracefully. This improves compatibility with different context types and ensures better tracing information.
 2026-03-10 11:23:29 +05:30
Aleix Conchillo Flaqué
d79e35d84f Add changelog for #3970 2026-03-09 20:47:42 -07:00
Aleix Conchillo Flaqué
00eb190424 Update daily-python to 0.24.0 2026-03-09 20:47:13 -07:00
Mark Backman
0dc95692ba Merge pull request #3967 from pipecat-ai/mb/fix-azure-stt-private-endpoint 2026-03-09 21:57:10 -04:00
Mark Backman
07b901c2a5 Add changelog for #3967 2026-03-09 15:05:20 -04:00
Mark Backman
f533dc3203 Fix Azure STT SpeechConfig failing when private_endpoint is provided 
SpeechConfig does not accept both `region` and `endpoint` simultaneously —
they are mutually exclusive. The previous code always passed both, which
raises ValueError when a user supplies a private_endpoint URL. Now we
conditionally pass either `endpoint` or `region`, never both.
 2026-03-09 15:05:20 -04:00
Paul Kompfner
20c3f553b2 Add missing 55-* update-settings examples for OpenPipe LLM and XTTS TTS 2026-03-09 14:36:15 -04:00
kompfner
02791cd503 Merge pull request #3965 from pipecat-ai/pk/fix-integration-tests 
Fix broken `test_unified_function_calling_anthropic` due to use of an…
 2026-03-09 13:42:35 -04:00
kompfner
f2debd9b1d Merge pull request #3963 from pipecat-ai/pk/improve-claude-changelog-skill 
Improve changelog skill: prioritize user-facing language and update e…
 2026-03-09 13:00:32 -04:00
kompfner
c0c49d0ddc Merge pull request #3964 from pipecat-ai/pk/add-some-missing-55-examples 
Add missing 55-* update-settings examples for Piper TTS, Kokoro TTS, …
 2026-03-09 12:59:36 -04:00
Mark Backman
3d1f866e73 Merge pull request #3951 from pipecat-ai/mb/remove-unused-imports-2026-03-07 
Remove unused imports, 2026-03-07
 2026-03-09 12:49:08 -04:00
Mark Backman
786279f143 Remove unused imports, 2026-03-07 2026-03-09 12:44:47 -04:00
Paul Kompfner
9423d22051 Fix broken test_unified_function_calling_anthropic due to use of an unsupported/deprecated model. 
Update the tests in test_integration_unified_function_calling.py to not specify particular models but instead just use service defaults (the tests shouldn't be model-dependent anyway)
 2026-03-09 12:07:56 -04:00
Paul Kompfner
f1bb065823 Add missing 55-* update-settings examples for Piper TTS, Kokoro TTS, Whisper STT, and Whisper MLX STT 
Also fix 13e-whisper-mlx.py to pass MLXModel.LARGE_V3_TURBO.value instead of the enum directly.
 2026-03-09 11:54:25 -04:00
Filipi da Silva Fuchter
0c5e936aa5 Merge pull request #3936 from pipecat-ai/filipi/fix_push_aggregation 
Fixed TTS context not being appended to the assistant message history
 2026-03-09 11:14:38 -04:00
filipi87
f0c5925a79 Fixing Piper test. 2026-03-09 12:07:45 -03:00
Paul Kompfner
7f9169269c Improve changelog skill: prioritize user-facing language and update example changelog 2026-03-09 10:45:33 -04:00
Mark Backman
c16e534f73 Merge pull request #3952 from pipecat-ai/mb/settings-alias 
Add Settings class attribute alias to all service classes
 2026-03-09 10:45:10 -04:00
filipi87
8ec160f71e Making the changelog more user friendly. 2026-03-09 11:37:11 -03:00
Filipi da Silva Fuchter
1e615cd095 Merge pull request #3962 from pipecat-ai/filipi/smallwebrtc_queue 
Queuing the messages received before the data channel is ready
 2026-03-09 10:29:05 -04:00
filipi87
ba87d1609c Only marking self._is_yielding_frames_synchronously if receiving TTSAudioRawFrame 2026-03-09 11:24:36 -03:00
Mark Backman
f7dc13c0de Update COMMUNITY_INTEGRATIONS.md for Settings alias class 2026-03-09 10:24:24 -04:00
filipi87
c5ce667387 Retrieving the context_id from the TTSStartedFrame 2026-03-09 11:10:42 -03:00
filipi87
097f9c0896 Fixed to push LLMAssistantPushAggregationFrame when the base TTSService class is responsible for pushing the TTSStoppedFrame. 2026-03-09 11:04:09 -03:00
Filipi da Silva Fuchter
16336a3ea4 Merge pull request #3937 from pipecat-ai/filipi/fix_orphan_function_call 
Fix context summarization leaving orphaned tool responses in kept context.
 2026-03-09 09:19:17 -04:00
Mark Backman
9eaa99c8e2 Merge pull request #3957 from pipecat-ai/mb/user-turn-completion-system-instruction 
Move turn completion instructions to system_instruction
 2026-03-09 09:17:06 -04:00
filipi87
4557ef8c42 Renaming method to _get_earliest_function_call_not_resolved_in_range 2026-03-09 10:16:02 -03:00
filipi87
aa693bb5ee Adding changelog entry for the SmallWebRTCConnection fix. 2026-03-09 10:11:40 -03:00
filipi87
74a06a6968 Adding extra comment. 2026-03-09 10:06:38 -03:00
filipi87
322e317a00 Adding guardrails in case the data channel is never established. 2026-03-09 10:04:33 -03:00