wx44wx/pipecat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
9,537 Commits 276 Branches 111 Tags
main
Commit Graph

939 Commits

Author SHA1 Message Date
joycech333
77cc314a08 feat: add Inception LLM service with Mercury-2 support 
Adds InceptionLLMService, an OpenAI-compatible service for Inception's
Mercury-2 diffusion-based reasoning model. Supports reasoning_effort
(instant/low/medium/high) and realtime mode for reduced TTFT.
 2026-05-21 11:23:23 -04:00
Mark Backman
87f5d60693 Merge pull request #4531 from pipecat-ai/mb/pipecat-prebuilt-1.0.1 
chore: bump pipecat-ai-prebuilt to 1.0.1
 2026-05-21 08:35:31 -04:00
Mark Backman
ee674ffb01 Add changelog for #4521 2026-05-20 17:57:43 -04:00
Mark Backman
105d6f27da Merge pull request #4514 from pipecat-ai/mb/websocket-stt-service-exception-handling 
Align websocket STT connection failures
 2026-05-20 15:15:35 -04:00
Mark Backman
a890ab7b21 Add changelog for PR #4531 2026-05-20 12:18:03 -04:00
Mark Backman
709a0ce839 Merge pull request #4527 from pipecat-ai/mb/fix-elevenlabs-keepalive-1008 
Fix ElevenLabs keepalive racing context-init (1008 disconnects)
 2026-05-20 11:21:17 -04:00
Mark Backman
be93350eae Merge pull request #4522 from pipecat-ai/mb/stt-latency-smallest 
Add P99 latency for Smallest AI, Mistral, XAI STT
 2026-05-20 11:21:00 -04:00
Mark Backman
4a96ab7073 Merge pull request #4524 from pipecat-ai/mb/fix-runner-imports 
Improve runner optional transport handling
 2026-05-20 11:16:16 -04:00
Mark Backman
70773bce0a Add changelog for PR #4527 2026-05-20 09:08:47 -04:00
filipi87
8bdb49bd1a chore: add changelogs for word-timestamp and frame-ordering fixes 2026-05-20 10:03:30 -03:00
Mark Backman
a5e6886b80 Fix ElevenLabs keepalive racing context-init (1008 disconnects) 
The keepalive could fire for a new turn's context before that context's
voice_settings context-init was sent, making the keepalive the context's
first message (no voice_settings) and causing ElevenLabs to reject the
later init with a 1008 policy violation. The keepalive now only targets a
context once its context-init has been sent (tracked in _context_init_sent).
 2026-05-20 08:59:01 -04:00
Mark Backman
82cd931efa Merge pull request #4306 from YFortin/fix/azure-tts-last-word-race 
fix(azure-tts): Route completion through word boundary queue to prevent last word from being missed
 2026-05-19 22:27:50 -04:00
Mark Backman
33e5d1f89b Add changelog for PR #4522 2026-05-19 18:33:58 -04:00
Mark Backman
861dd23873 Add changelog for runner updates 2026-05-19 17:31:07 -04:00
Mark Backman
e298491068 Add changelog for websocket STT failure handling 2026-05-18 12:41:56 -04:00
asilvestre
cab4585cbb added changelog 2026-05-18 14:40:56 +02:00
Filipi da Silva Fuchter
c51a817efa Merge pull request #4442 from pipecat-ai/filipi/runner_all_transports 
Unified start route to make all transports available
 2026-05-18 09:27:44 -03:00
Bismeet singh
d85eda6da8 Merge pull request #4507 from BismeetSingh/fix/elevenlabs-stt-service-crash-language 
Fix/elevenlabs stt service crash language
 2026-05-17 10:17:07 -04:00
aconchillo
6b93ca0cb6 Update changelog for version 1.2.1 2026-05-15 22:18:46 +00:00
Aleix Conchillo Flaqué
b6ecce754b Merge pull request #4501 from pipecat-ai/aleix/fix-filter-incomplete-tool-calls 
Fix filter-incomplete + function-calling deadlock
 2026-05-15 15:11:45 -07:00
Aleix Conchillo Flaqué
d39e6bf921 Add changelog for #4501 2026-05-15 14:54:51 -07:00
Mark Backman
58a22aeeb1 Add changelog for #4500 2026-05-15 15:19:39 -04:00
aconchillo
b13af2b053 Update changelog for version 1.2.0 2026-05-14 21:45:36 +00:00
Aleix Conchillo Flaqué
8e405f15aa changelog: fix 4446.change.md file name 2026-05-14 14:38:54 -07:00
Aleix Conchillo Flaqué
44a40e8eb2 Merge pull request #4497 from pipecat-ai/aleix/fix-tts-context-id-fallback 
Fall back to _turn_context_id in get_active_audio_context_id
 2026-05-14 13:34:34 -07:00
Aleix Conchillo Flaqué
ea97cb1a78 Add changelog for #4497 2026-05-14 13:22:50 -07:00
Mark Backman
dcbb0070c9 Add changelog for Soniox language selection 2026-05-14 15:42:43 -04:00
filipi87
c8efe319b3 Adding the changelog for the changes. 2026-05-14 11:10:33 -03:00
Mark Backman
49bda11ae8 Merge pull request #4482 from pipecat-ai/mb/soniox-stt-token-language 
Propagate Soniox token language
 2026-05-13 16:28:56 -04:00
Aleix Conchillo Flaqué
07640582ce Merge pull request #4467 from pipecat-ai/aleix/fix-tts-ttfb-tracing 
Fix metrics.ttfb and partial output on TTS/STT/LLM OpenTelemetry spans
 2026-05-13 13:10:52 -07:00
Mark Backman
078af6969a Merge pull request #4473 from timofey-TK/inworld-tts-v2 
Add support for Inworld TTS v2 fields
 2026-05-13 15:32:16 -04:00
Mark Backman
9f40ba21c2 Add changelog for Soniox language fix 2026-05-13 15:26:10 -04:00
kompfner
7e4cd23de4 Merge pull request #4474 from pipecat-ai/pk/inworld-realtime-tools 
Extend cancel_on_interruption=False to Inworld Realtime (best-effort + warning)
 2026-05-13 15:12:34 -04:00
TimTk
97f50c8aa2 Address review: use resolve_language, narrow delivery_mode type, update changelog 
- Replace custom LANGUAGE_MAP fallback in language_to_inworld_language with
  resolve_language(language, LANGUAGE_MAP, use_base_code=False) to match the
  pattern used by other services and restore the unverified-language warning
- Tighten delivery_mode type from str to Literal["STABLE", "BALANCED", "CREATIVE"]
- Update changelog entry to mention delivery_mode and language normalization
 2026-05-13 21:43:02 +03:00
Mark Backman
08680732f6 Merge pull request #4475 from pipecat-ai/mb/cartesia-korean-fix 
Fix Cartesia CJK timestamp spacing
 2026-05-13 13:20:42 -04:00
Mark Backman
064b68aa01 Fix Cartesia CJK timestamp spacing 2026-05-13 13:13:40 -04:00
Mark Backman
5fef239b68 Merge pull request #4450 from pipecat-ai/mb/gpt-realtime-whisper 
Default OpenAI Realtime transcription to gpt-realtime-whisper
 2026-05-13 09:48:33 -04:00
Filipi da Silva Fuchter
9148e307cc Merge pull request #4464 from pipecat-ai/filipi/nvidia_sagemaker 
NVidia sagemaker - TTS and STT services
 2026-05-13 07:53:26 -03:00
Aleix Conchillo Flaqué
52d8008783 Add LLM interruption changelog entry for #4467 2026-05-12 20:10:43 -07:00
Aleix Conchillo Flaqué
e70ee603b2 Add STT changelog entry for #4467 2026-05-12 20:10:43 -07:00
Aleix Conchillo Flaqué
079282d140 Add changelog for #4467 2026-05-12 20:10:43 -07:00
filipi87
0740021ff4 Removing changelog for sanitize_text_for_tts 2026-05-12 18:29:35 -03:00
Mark Backman
2fdb9bbf42 Merge pull request #4462 from pipecat-ai/mb/cartesia-sonic-3.5 2026-05-12 16:34:04 -04:00
Paul Kompfner
863a1bf177 Add changelog for #4474 2026-05-12 16:04:12 -04:00
Paul Kompfner
58333b2705 Extend cancel_on_interruption=False to InworldRealtimeLLMService (best-effort) 
Same async-tool routing approach as #4441: detect async-tool messages in
the LLM context, deliver the final result via the formal tool-result
channel.

Caveat: as of this writing, Inworld Realtime doesn't appear to handle
the resulting delayed tool result reliably, so the routing is
best-effort and the service emits a one-time warning when async-tool
messages are seen. Streamed intermediate results remain unsupported.

Also adds function calling to the realtime-inworld.py example, and
softens the Inworld mention in the #4447 changelog now that the
exclusion is being closed.
 2026-05-12 16:03:34 -04:00
TimTk
ecaff1d1eb Fix changelog fragment number 2026-05-12 22:21:59 +03:00
Mark Backman
e2bfa6352f Add changelog for #4450 2026-05-12 15:20:57 -04:00
TimTk
9b55d4ddd4 Add support for Inworld TTS v2 fields 2026-05-12 22:13:09 +03:00
Paul Kompfner
fc0589e8f1 Switch OpenAIRealtimeLLMService default model to gpt-realtime-2 2026-05-12 14:57:59 -04:00
Paul Kompfner
a52bdef32b Add reasoning support to OpenAIRealtimeLLMService for gpt-realtime-2 2026-05-12 13:55:19 -04:00