Commit Graph

  • 1c88565725 Merge pull request #2772 from pipecat-ai/mb/fix-openai-realtime-import Mark Backman 2025-10-02 12:34:16 -04:00
  • 07a6c2fb0e Merge pull request #2775 from pipecat-ai/aleix/pipecat-0.0.87 Aleix Conchillo Flaqué 2025-10-02 09:12:41 -07:00
  • e99f3bf75a update CHANGELOG for 0.0.87 Aleix Conchillo Flaqué 2025-10-02 09:09:09 -07:00
  • f09d780413 Fix: Resolve docstring build issues before 0.0.87 release Mark Backman 2025-10-02 10:09:25 -04:00
  • e370d23374 Fix: Change import for OpenAIRealtimeLLMContext in OpenAIRealtimeLLMService Mark Backman 2025-10-02 09:39:44 -04:00
  • b68ec14146 Updates to publish workflows Mark Backman 2025-10-02 08:25:35 -04:00
  • c219f8d55d Add WebSocket reconnection test example hush/reset-closing-flag James Hush 2025-10-02 14:24:09 +08:00
  • 45ebd042ac Fix: Reset _closing flag on WebSocket reconnection to ensure on_client_disconnected fires James Hush 2025-10-02 14:11:14 +08:00
  • c567fd71b1 Merge pull request #2747 from pipecat-ai/filipi/whatsapp_runner Filipi da Silva Fuchter 2025-10-01 21:21:34 -03:00
  • 2ca1b2d6f8 Merge pull request #2612 from pipecat-ai/filipi/deepgram_flux Filipi da Silva Fuchter 2025-10-01 21:20:47 -03:00
  • 04041a9a9a Merge pull request #2757 from pipecat-ai/hush/retryTimeout Mark Backman 2025-10-01 19:08:09 -04:00
  • 6c498dc70f Merge pull request #2745 from pipecat-ai/aleix/transport-message-frames-deprecations Aleix Conchillo Flaqué 2025-10-01 16:05:55 -07:00
  • 32b07c1720 Fix AWS Bedrock timeout exception handling James Hush 2025-09-30 15:31:33 +08:00
  • ad507ce23d FrameLogger: it's fine to print transport messages Aleix Conchillo Flaqué 2025-10-01 15:29:30 -07:00
  • be562cedfc DailyTransport: deprecate DailyTransportMessage(Urgent)Frame Aleix Conchillo Flaqué 2025-09-25 14:23:39 -07:00
  • 089e703e1f LiveKitTransport: deprecate LiveKitTransportMessage(Urgent)Frame Aleix Conchillo Flaqué 2025-09-25 14:22:34 -07:00
  • 4dc1e15a99 frames: use OutputTransportMessage(Urgent)Frame instead of TransportMessage(Urgent)Frame Aleix Conchillo Flaqué 2025-09-25 13:04:31 -07:00
  • c7dc2e886f frames: use InputTransportMessageFrame instead of InputTransportMessageUrgentFrame Aleix Conchillo Flaqué 2025-09-25 12:07:31 -07:00
  • 11bc4ea854 Adding deepgram flux to release evals. Filipi Fuchter 2025-10-01 19:24:58 -03:00
  • 029d76033d Merge pull request #2765 from pipecat-ai/mb/remove-daily-logging-04a Mark Backman 2025-10-01 17:52:33 -04:00
  • 924d7dea9a Merge pull request #2766 from pipecat-ai/aleix/rtvi-properly-deprecate-errors-enabled Aleix Conchillo Flaqué 2025-10-01 14:49:12 -07:00
  • 244e94f3ce RTVIParams: properly deprecate errors_enabled Aleix Conchillo Flaqué 2025-10-01 14:30:41 -07:00
  • af1f51d49e Remove DailyLogLevel from 04a example Mark Backman 2025-10-01 17:06:35 -04:00
  • 9ba3c168b8 Merge pull request #2756 from pipecat-ai/filipi/esp32 Filipi da Silva Fuchter 2025-10-01 16:05:47 -03:00
  • e6ee8f7a16 New example using DeepgramFluxSTTService. Filipi Fuchter 2025-10-01 15:43:25 -03:00
  • 2ea2bd99e0 Deepgram Flux speech-to-text service implementation. Filipi Fuchter 2025-10-01 15:43:09 -03:00
  • 0c2ced7c52 Created WebsocketSTTService base class. Filipi Fuchter 2025-10-01 15:42:56 -03:00
  • fb160646b8 Fixing the SDP munging to keep it working on Chrome. Filipi Fuchter 2025-10-01 14:18:39 -03:00
  • 0dac7f7e48 Update OpenAIRealtimeLLMService to work with LLMContext and LLMContextAggregatorPair pk/openai-realtime-universal-llmcontext Paul Kompfner 2025-10-01 10:18:07 -04:00
  • a2c69fbd8b Update OpenAIRealtimeLLMService to work with LLMContext and LLMContextAggregatorPair Paul Kompfner 2025-10-01 09:57:51 -04:00
  • 89fed57af2 Merge pull request #2748 from pipecat-ai/filipi/remove_smallwebrtc_queue Filipi da Silva Fuchter 2025-10-01 08:07:47 -03:00
  • a39f8b4882 Remove extra code hush/prerecordedMessageExample James Hush 2025-10-01 14:48:12 +08:00
  • 76fc36f621 Pre recorded message example James Hush 2025-10-01 14:43:36 +08:00
  • c0878c5e09 Save progress James Hush 2025-10-01 13:58:32 +08:00
  • c6a1013051 Pre-recorded message example James Hush 2025-10-01 13:52:58 +08:00
  • ae8b9f0756 Fix: Ensure on_client_disconnected event always fires in FastAPIWebsocketTransport fix/fastapi-websocket-disconnect-event James Hush 2025-10-01 10:15:46 +08:00
  • feae3b6d2d Merge pull request #2742 from pipecat-ai/aleix/deprecate-daily-update-remote-participants-frame Aleix Conchillo Flaqué 2025-09-30 16:27:34 -07:00
  • 92d3be8975 DailyTransport: deprecated DailyUpdateRemoteParticipantsFrame Aleix Conchillo Flaqué 2025-09-25 09:38:14 -07:00
  • 0f53e1db2c Merge pull request #2759 from pipecat-ai/aleix/dont-cancel-if-finished Aleix Conchillo Flaqué 2025-09-30 16:21:16 -07:00
  • d398e8cc10 Merge pull request #2761 from pipecat-ai/aleix/rtvi-tail-updates Aleix Conchillo Flaqué 2025-09-30 13:55:17 -07:00
  • e5f263d380 update CHANGELOG Aleix Conchillo Flaqué 2025-09-30 13:51:35 -07:00
  • 3a4c303c54 RTVIParams: add errors_enabled deprecation warnings Aleix Conchillo Flaqué 2025-09-30 13:49:51 -07:00
  • 54a1ef47d0 Merge pull request #2758 from pipecat-ai/mb/claude-sonnet-4.5 Mark Backman 2025-09-30 16:42:47 -04:00
  • 149ffa4f3c RTVIObserver: add support system logs Aleix Conchillo Flaqué 2025-09-30 13:40:48 -07:00
  • e5465034d9 RTVIObserver: add support for user/bot audio levels Aleix Conchillo Flaqué 2025-09-30 13:37:08 -07:00
  • 568c7c782d rtvi: allow None RTVIProcessor and rename to send_rtvi_message() Aleix Conchillo Flaqué 2025-09-30 13:35:27 -07:00
  • 9851334221 rtvi: deprecate errors_enabled and always send errors Aleix Conchillo Flaqué 2025-09-30 13:31:09 -07:00
  • e79c4fc99d PipelineTask: avoid cancellation if application is finished Aleix Conchillo Flaqué 2025-09-30 13:18:25 -07:00
  • 55c321f4ff Merge pull request #2751 from pipecat-ai/aleix/nova-sonic-disconnect-fix Aleix Conchillo Flaqué 2025-09-30 13:12:22 -07:00
  • 8962263329 Update OpenAIRealtimeLLMService to work with LLMContext and LLMContextAggregatorPair Paul Kompfner 2025-09-30 14:47:06 -04:00
  • ae22673595 Update OpenAIRealtimeLLMService to work with LLMContext and LLMContextAggregatorPair Paul Kompfner 2025-09-30 10:52:28 -04:00
  • 236ac93ac6 Remove remaining usage of OpenAILLMContext throughout the codebase in favor of LLMContext, except for: - Usage in classes that are already deprecated - Usage related to realtime LLMs, which don't yet support LLMContext - Usage in (soon-to-be-deprecated) code paths related to OpenAILLMContext itself and associated machinery Paul Kompfner 2025-09-24 15:54:01 -04:00
  • a14a53a005 Merge pull request #2735 from pipecat-ai/pk/remove-openaillmcontext-usage kompfner 2025-09-30 10:09:25 -04:00
  • a71f937e8f Update AnthropicLLMService to use claude-sonnet-4-5-20250929 Mark Backman 2025-09-30 08:49:30 -04:00
  • ceba27e696 Address some AWSNovaSonicLLMService context-recording edge cases Paul Kompfner 2025-09-29 11:46:56 -04:00
  • 032032df65 Only remove ESP32 ICE candidates if host is defined. Filipi Fuchter 2025-09-29 15:42:23 -03:00
  • 07ba02a491 Get rid of LLMContext.get_messages_for_persistent_storage(). Paul Kompfner 2025-09-29 10:56:00 -04:00
  • 4d9873b613 Update AWSNovaSonicLLMService to work with LLMContext and LLMContextAggregatorPair Paul Kompfner 2025-09-25 09:46:33 -04:00
  • d0178edad0 Merge pull request #2753 from pipecat-ai/mb/quickstart-0.0.86 Mark Backman 2025-09-29 09:43:33 -04:00
  • 795c5e55d9 Quickstart: Update to 0.0.86, removing pytorch requirements Mark Backman 2025-09-27 08:30:37 -04:00
  • 8f8d8ae0d8 AWSNovaSonicLLMService: add missing await Aleix Conchillo Flaqué 2025-09-26 15:58:05 -07:00
  • 741f192d04 Merge pull request #2096 from pipecat-ai/vp-mcp-ex-nit Vanessa Pyne 2025-09-26 10:21:22 -05:00
  • a5595b82ea removing the message queue inside the SmallWebRTCConnection. Filipi Fuchter 2025-09-26 11:02:17 -03:00
  • 4d1915eb41 Fixing ruff format. Filipi Fuchter 2025-09-26 10:49:52 -03:00
  • b3a84fc772 Refactoring how we are handling the lifespan inside the runner. Filipi Fuchter 2025-09-26 10:47:04 -03:00
  • 403d22e62c Creating the whatsapp routes inside the runner. Filipi Fuchter 2025-09-26 10:28:19 -03:00
  • ee00ee5c57 Merge pull request #2744 from pipecat-ai/aleix/vad-analyzer-thread-executor Aleix Conchillo Flaqué 2025-09-25 13:43:34 -07:00
  • f53fd880dc BaseInputTransport: create VAD thread in VADAnalyzer Aleix Conchillo Flaqué 2025-09-25 11:37:16 -07:00
  • de3461e4cc Merge pull request #2743 from pipecat-ai/aleix/turn-analyzer-fixes Aleix Conchillo Flaqué 2025-09-25 13:40:43 -07:00
  • 7bafc3a1bb BaseSmartTurn: process speech in a separate thread Aleix Conchillo Flaqué 2025-09-25 11:12:25 -07:00
  • 22ef61fe8d BaseTurnAnalyzer: add BaseTurnParams base class for parameters Aleix Conchillo Flaqué 2025-09-25 09:47:58 -07:00
  • 7078fb53bd Merge pull request #2738 from pipecat-ai/aleix/openai-cached-tokens-metrics Aleix Conchillo Flaqué 2025-09-25 13:36:03 -07:00
  • 0a163201ea feat: Add sentence aggregation and Whisker debugger to transcript processor hush/aggregateTranscripts James Hush 2025-09-25 14:01:19 +08:00
  • 33447ad6f2 BaseOpenAILLMService: include cached tokens to metrics frame Aleix Conchillo Flaqué 2025-09-24 19:32:16 -07:00
  • 6faa50ae5b Remove remaining usage of OpenAILLMContext throughout the codebase in favor of LLMContext, except for: - Usage in classes that are already deprecated - Usage related to realtime LLMs, which don't yet support LLMContext - Usage in (soon-to-be-deprecated) code paths related to OpenAILLMContext itself and associated machinery Paul Kompfner 2025-09-24 15:54:01 -04:00
  • 3797f41c8c Merge pull request #2734 from pipecat-ai/aleix/pipecat-0.0.86 v0.0.86 Aleix Conchillo Flaqué 2025-09-24 12:19:16 -07:00
  • ff919b8c15 update CHANGELOG for 0.0.86 Aleix Conchillo Flaqué 2025-09-24 11:28:14 -07:00
  • cb048d6c7e Merge pull request #2733 from pipecat-ai/aleix/tavus-missing-daily-callback Aleix Conchillo Flaqué 2025-09-24 11:10:32 -07:00
  • 6c2c43ade0 Merge pull request #2724 from pipecat-ai/pk/update-natural-conversation-examples-with-universal-context Aleix Conchillo Flaqué 2025-09-24 11:07:50 -07:00
  • f899c15b03 Merge pull request #2731 from pipecat-ai/pk/update-example-25-to-use-universal-context Aleix Conchillo Flaqué 2025-09-24 11:05:36 -07:00
  • d10ef08775 Merge pull request #2727 from pipecat-ai/pk/strands-agents-needs-to-support-openaillmcontextframe-for-now Aleix Conchillo Flaqué 2025-09-24 11:05:07 -07:00
  • 27a5af6fa1 Merge pull request #2728 from pipecat-ai/pk/fix-playht-in-env-example Aleix Conchillo Flaqué 2025-09-24 11:04:55 -07:00
  • 4bff0a7c49 Merge pull request #2732 from pipecat-ai/pk/update-voicemail-detector-to-use-llm-context Aleix Conchillo Flaqué 2025-09-24 11:04:42 -07:00
  • 508f7d203d Merge pull request #2729 from pipecat-ai/aleix/frame-processor-cancel-default-timeout Aleix Conchillo Flaqué 2025-09-24 10:55:52 -07:00
  • 0f87d5342c Merge pull request #2266 from pipecat-ai/filipi/hey_gen_transport Filipi da Silva Fuchter 2025-09-24 14:35:50 -03:00
  • f6164e3bde HeyGen implementation for Pipecat - HeyGenTransport. Filipi Fuchter 2025-09-24 14:33:56 -03:00
  • 1a0fb55d0f TavusTransport: add missing on_before_leave callback Aleix Conchillo Flaqué 2025-09-24 10:18:56 -07:00
  • 6d0beef944 Update VoicemailDetector to use universal LLMContext Paul Kompfner 2025-09-24 12:58:42 -04:00
  • b9fd6b873b Update comment in example 07b to reference LLMContext rather than OpenAILLMContext Paul Kompfner 2025-09-24 12:49:34 -04:00
  • dea0f1791f Update OpenAILLMAdapter.get_messages_for_logging() to truncate "input_audio" message data Paul Kompfner 2025-09-24 12:41:41 -04:00
  • da66c38795 Update example 25 to use universal LLMContext Paul Kompfner 2025-09-24 12:37:29 -04:00
  • 912f8b96f0 Fix PlayHT env variable names in env.example Paul Kompfner 2025-09-24 11:24:35 -04:00
  • f9eb447d82 FrameProcessor: timeout when cancelling tasks Aleix Conchillo Flaqué 2025-09-24 08:24:28 -07:00
  • 65f5fe8588 StrandsAgentsProcessor should still support OpenAILLMContextFrame until that frame has been deprecated Paul Kompfner 2025-09-24 11:05:27 -04:00
  • 817c77f3fe Update SmallWebRTCTransport to pass a sender ID in the "on_app_message" event Paul Kompfner 2025-09-24 10:24:59 -04:00
  • 8896179b00 Update another "natural conversation" example to use universal LLMContext. Note that this one had to also be fixed in various ways, as it wasn't working. Paul Kompfner 2025-09-24 09:55:11 -04:00
  • 463752360b Merge pull request #2726 from pipecat-ai/mb/twilio-serializer-cleanup Filipi da Silva Fuchter 2025-09-24 10:45:38 -03:00
  • 66b7977a62 Make SmallWebRTCTransport adhere to the expected "on_app_message" event signature Paul Kompfner 2025-09-24 09:34:28 -04:00
  • 468de68aec Fixup for TwilioFrameSerializer Mark Backman 2025-09-24 09:32:46 -04:00
  • c4762c1a92 Merge pull request #2627 from jessieweiyi/main Mark Backman 2025-09-24 09:13:10 -04:00