Skip to content

Pull requests: pipecat-ai/pipecat

New pull request New
155 Open 4,016 Closed
155 Open 4,016 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Add AssemblyAI Sync speech-to-text service
#5488 opened Aug 29, 2026 by gsharp-aai Loading…
fix(vad): evaluate the VAD thresholds per analysis frame and honor speech_activity_period
#5486 opened Aug 28, 2026 by Kayvan-Zahiri Loading…
Expose actual OpenAI response service tiers in usage metrics
#5485 opened Aug 28, 2026 by johnparkdd Contributor Loading…
1
Fix the start of the bot's first utterance being lost on MoQ cold starts
#5484 opened Aug 28, 2026 by vipyne Member • Draft
1
Adds an opt-out for GeminiSTTService's VAD-triggered audio_stream_end flush, which truncates utterances when a speaker pauses mid-sentence. Default is unchanged. No existing issue — found while measuring STT latency in a production voice agent
#5483 opened Aug 28, 2026 by arnavXgupta Loading…
1
fix(deepgram-flux): reconnect for connection-only settings and surface FatalError details
#5477 opened Aug 28, 2026 by dakshdua Contributor Loading…
1
Feat/fish audio stt
#5476 opened Aug 28, 2026 by Thileepan Loading…
Play an audio file as a user turn in evals
#5473 opened Aug 28, 2026 by aconchillo Contributor Loading…
1
fix(openai): scope Responses reasoning context
#5467 opened Aug 27, 2026 by timofey-TK Contributor Loading…
1
1
fix(asyncai): apply model, voice and language changes by opening a new session
#5464 opened Aug 27, 2026 by Rehansanjay Loading…
Keep muted stop frames from stranding user speaking state
#5455 opened Aug 26, 2026 by arnavXgupta Loading…
fix(turns): clear speaking state on muted stop frames
#5453 opened Aug 26, 2026 by sir-ad Loading…
Break the latency breakdown into contributions covering the whole turn
#5445 opened Aug 26, 2026 by markbackman Contributor • Draft
3
Retire the MoQ audio track on interruption
#5443 opened Aug 26, 2026 by kixelated Contributor • Draft
1
feat: add Synthorai LLM service
#5440 opened Aug 26, 2026 by cuihuan Loading…
Add a normalization setting to the Cartesia TTS services provider-watch Opened by the provider-watch bot
#5430 opened Aug 25, 2026 by pipecat-provider-watch-bot Bot • Draft
2
fix(context): keep whole tags as single tokens when diffing TTS text
#5421 opened Aug 24, 2026 by sohumh Contributor • Draft
1
fix(context): emit a final progress frame when force-completing a spoken slot
#5420 opened Aug 24, 2026 by sohumh Contributor • Draft
1
fix(context): match word-timestamp tokens that carry leading punctuation
#5419 opened Aug 24, 2026 by sohumh Contributor • Draft
1
feat(services/qwen): add Qwen3STTService for local speech-to-text
#5415 opened Aug 24, 2026 by RoKenobi Loading…
1 of 2 tasks
1
fix(gladia): drain final transcripts on stop
#5406 opened Aug 22, 2026 by dorukdumlu Contributor Loading…
1
1
Resolve the ElevenLabs STT detected language to a Language enum
#5403 opened Aug 21, 2026 by seygi Loading…
1
fix(soniox): drain final transcripts on stop
#5402 opened Aug 21, 2026 by nateschickler0 Loading…
1
2
test(observers): buy the startup span assertion its headroom from the count
#5400 opened Aug 21, 2026 by dylanpulver Contributor Loading…
1
Add optional prompt_cache_key passthrough for OpenAI LLM services
#5391 opened Aug 21, 2026 by mannyb223 Loading…
ProTip! no:milestone will show everything without a milestone.