{"full_name":"pipecat-ai/pipecat","owner":"pipecat-ai","name":"pipecat","shard":"1","url":"https://github.com/pipecat-ai/pipecat","description":"Open Source framework for voice agents, multimodal apps, and realtime AI. Maintained by Daily and the community.","homepage":"https://pipecat.ai","topics":["ai","real-time","voice","voice-assistant","chatbot-framework","chatbots"],"tag_pages":["ai","chatbots","real-time","voice","voice-assistant"],"language":"Python","license":"BSD-2-Clause","avatar":"https://avatars.githubusercontent.com/u/169612734?v=4","created_at":"2023-12-27T12:59:00Z","pushed_at":"2026-11-15T00:30:48Z","metrics":{"stars":17271,"forks":3117,"watchers":85,"issues":1482,"prs":5321,"releases":126,"issues_open":137,"issues_closed":1345,"prs_open":348,"prs_closed":1016,"prs_merged":3957,"discussions":0},"growth":{"day":{"stars":10,"forks":6,"base_day":"2026-11-15"}},"history":{"dates":["2026-11-10","2026-11-11","2026-11-12","2026-11-13","2026-11-14","2026-11-15","2026-11-16"],"stars":[17170,17199,17219,17224,17247,17261,17271],"forks":[3078,3090,3100,3101,3108,3111,3117],"watchers":[null,83,83,84,85,85,85],"issues_open":[null,129,133,135,134,135,137],"prs_open":[null,318,336,337,338,344,348]},"boards":{},"same_owner":[],"issues":[{"number":3989,"title":"Allow excluding specific messages from context summarization","comments":9,"created_at":"2026-03-11T10:56:38Z"},{"number":1015,"title":"Add WordTTSService support for aws PollyTTSService","comments":9,"created_at":"2025-01-16T17:34:52Z"},{"number":5683,"title":"TranscriptionFrame queued behind the frame that starts a user turn is dropped by that turn's own interruption","comments":7,"created_at":"2026-09-10T14:17:00Z"},{"number":6110,"title":"AnthropicLLMService: after a tool or system prompt change, every later request fails with a 400 on accounts with preserved thinking enforced","comments":6,"created_at":"2026-10-11T04:21:19Z"},{"number":4996,"title":"Interruption loses played TTS text from assistant context (InterruptionFrame overtakes released TTSTextFrames + transport queue drop) — bot repeats itself","comments":6,"created_at":"2026-07-10T00:22:17Z"},{"number":3985,"title":"Proposal: graceful interruption release for TTS/output transports","comments":6,"created_at":"2026-03-11T03:07:54Z"},{"number":2811,"title":"Option to Initiate Outbound Whatsapp Business Call","comments":6,"created_at":"2025-10-08T02:26:17Z"},{"number":1882,"title":"Support for combined STT + LLM with gpt4o-audio-preview","comments":6,"created_at":"2025-05-24T04:41:58Z"}],"discussions":[],"latest_release":{"name":"v1.16.0","tag":"v1.16.0","published_at":"2026-11-12T00:45:03Z","body":"### Added\n\n- Added a `normalization` field to `CartesiaTTSService.Settings` and `CartesiaHttpTTSService.Settings`, passing Cartesia's text normalization mode (`\"auto\"`, `\"off\"`, or a locale code such as `\"en-GB\"`) through to synthesis requests.\n  (PR [#5430](https://github.com/pipecat-ai/pipecat/pull/5430))\n\n- Added `SarvamHttpSTTService`, which sends each VAD speech segment to Sarvam's REST speech-to-text API (`saaras:v4` by default, or `saaras:v3`) and yields the transcription from `run_stt()`. It takes the same `mode` and `keyterms` options as `SarvamSTTService`, and sends a segment longer than the API's 30-second limit as consecutive pieces whose transcripts are joined.\n\n    ```python\n    stt = SarvamHttpSTTService(\n        api_key=os.getenv(\"SARVAM_API_KEY\"),\n        settings=SarvamHttpSTTService.Settings(language=Language.HI_IN),\n    )\n    ```\n  (PR [#5501](https://github.com/pipecat-ai/pipecat/pull/5501))\n\n- Added `max_speakers` and `filter_profanity` to `AssemblyAISTTService.Settings`, passing AssemblyAI's speaker-label cap and profanity filter through to the streaming session.\n  (PR [#5530](https://github.com/pipecat-ai/pipecat/pull/5530))\n\n- `CartesiaSTTService.Settings` now exposes Cartesia's `min_volume` and `max_silence_duration_secs`, which control the silence that triggers automatic transcript finalization on ink-whisper models.\n  (PR [#5531](https://github.com/pipecat-ai/pipecat/pull/5531))\n\n- Added `speed` to Inworld Realtime's `AudioOutput`, so a session can set the agent's speaking rate.\n  (PR [#5541](https://github.com/pipecat-ai/pipecat/pull/5541))\n\n- Added `noise_reduction` to Inworld Realtime's `AudioInput`, so a session can enable Inworld's near-field or far-field input filter.\n  (PR [#5541](https://github.com/pipecat-ai/pipecat/pull/5541))\n\n- Added `threshold`, `prefix_padding_ms`, `silence_duration_ms` and `idle_timeout_ms` to Inworld Realtime's `TurnDetection`, so `type=\"server_vad\"` sessions can be tuned the way Inworld documents.\n  (PR [#5541](https://github.com/pipecat-ai/pipecat/pull/5541))\n\n- Added `language` and `prompts` to Inworld Realtime's `InputTranscription`, so a session can pin the STT language and list expected words or phrases for the transcriber to recognize.\n  (PR [#5541](https://github.com/pipecat-ai/pipecat/pull/5541))\n\n- Added an `instruction` field to `InworldHttpTTSService.Settings`, which passes Inworld's plain-English steering for a request (e.g. `\"speak loudly and urgently\"`) through to `inworld-tts-2`. The WebSocket API takes steering as inline `[tags]` in the text instead, so `InworldTTSService` ignores the field.\n  (PR [#5542](https://github.com/pipecat-ai/pipecat/pull/5542))\n\n- Added `pronunciation_dict` to `MiniMaxHttpTTSService.Settings`, passing MiniMax's `pronunciation_dict.tone` replacements (e.g. `\"Pipecat/(ˈpaɪpkæt)\"`) through to synthesis requests.\n  (PR [#5545](https://github.com/pipecat-ai/pipecat/pull/5545))\n\n- Added a `truncation` field to the OpenAI Realtime `SessionProperties`, which sets what the server does once a long session outgrows the model's input token limit. By default it drops the oldest messages, only as many as each response needs. `TruncationRetentionRatio` drops more at once so truncation happens less often and more input stays in the prompt cache, which OpenAI recommends for lowering the cost of long sessions; `\"disabled\"` returns an error instead of truncating.\n  (PR [#5547](https://github.com/pipecat-ai/pipecat/pull/5547))\n\n- Added `delay`, `keywords` and `languages` to the OpenAI Realtime `InputAudioTranscription` settings, exposing OpenAI's input-transcription latency/accuracy control and transcription hints.\n  (PR [#5547](https://github.com/pipecat-ai/pipecat/pull/5547))\n\n- Added `speaker_id`, `length_scale`, `noise_scale` and `noise_w_scale` settings to `PiperTTSService` and `PiperHttpTTSService`, exposing Piper's speaker selection, speaking speed and variability controls.\n  (PR [#5551](https://github.com/pipecat-ai/pipecat/pull/5551))","is_prerelease":false},"readme":true,"readme_zh":false,"news":[],"stale":false,"snapshot_at":"2026-11-16T09:23:14Z","history_start":"2026-09-06"}