Commit Graph

3011 Commits

Author SHA1 Message Date
Aleix Conchillo Flaqué
b61846534d SpeechmaticsSTTService: improve exception handling and loggin 2025-08-21 18:42:23 -07:00
Aleix Conchillo Flaqué
24a628c85e remove watchdog timers and specific asyncio implementations
Watchdog timers have been removed. They were introduced in 0.0.72 to help
diagnose pipeline freezes. Unfortunately, they proved ineffective since they
required developers to use Pipecat-specific queues, iterators, and events to
correctly reset the timer, which limited their usefulness and added friction.
2025-08-21 18:09:56 -07:00
Mark Backman
cb13f4b4cb Add user speaking and transcription frames to idle_timeout_frames 2025-08-21 06:43:10 -04:00
Aleix Conchillo Flaqué
28c729cc36 replace asyncio.wait_for for wait_for2.wait_for 2025-08-20 15:26:57 -07:00
Aleix Conchillo Flaqué
4ff0567025 BaseObject: allow keyword arguments 2025-08-20 15:14:31 -07:00
Aleix Conchillo Flaqué
1377dec01b DTMFAggregator: no need for interruption task
Now that system frames are queued there's no need to have an additional task to
push a `BotInterruptionFrame`.
2025-08-20 14:35:04 -07:00
Aleix Conchillo Flaqué
f1c1ebf852 timeout: fix wait_for2 import 2025-08-20 14:24:16 -07:00
Aleix Conchillo Flaqué
f387776985 add custom asyncio.wait_for()
This patch uses `wait_for2` package to implement `asyncio.wait_for()` for
Python < 3.12.

In Python 3.12, `asyncio.wait_for()` is implemented in terms of
`asyncio.timeout()` which fixed a bunch of issues. However, this was never
backported (because of the lack of `async.timeout()`) and there are still many
remainig issues, specially in Python 3.10, in `async.wait_for()`.

See https://github.com/python/cpython/pull/98518
2025-08-20 14:09:05 -07:00
Aleix Conchillo Flaqué
6831e63ec9 PipelineTask: use PipelineSource/PipelineSink and remove tasks 2025-08-20 10:08:54 -07:00
Aleix Conchillo Flaqué
12bcb7db64 ParallelPipeline: use PipelineSource/PipelineSink and remove tasks 2025-08-20 10:08:54 -07:00
Aleix Conchillo Flaqué
1b48b1d860 Pipeline: allow passing user source and sink processors 2025-08-20 10:08:54 -07:00
Aleix Conchillo Flaqué
d161e2767f FrameProcessor: allow pausing/resuming system frames 2025-08-20 10:08:54 -07:00
Aleix Conchillo Flaqué
4015aedb86 tests: fix unit tests 2025-08-20 10:08:54 -07:00
Aleix Conchillo Flaqué
75a6ee839b BaseObserver: added new on_process_frame 2025-08-20 10:08:54 -07:00
Aleix Conchillo Flaqué
13ce02c896 FrameProcessor: add new entry_processors() method 2025-08-20 10:08:54 -07:00
Aleix Conchillo Flaqué
2fd5885dc3 pipeline: implement processors property 2025-08-20 07:40:21 -07:00
Aleix Conchillo Flaqué
d743586bfb BasePipeline: move processors_with_metrics() to FrameProcessor 2025-08-20 07:40:21 -07:00
Aleix Conchillo Flaqué
8051017895 pipeline: wrap with pipelines, use direct mode and reduce tasks 2025-08-20 07:40:21 -07:00
Aleix Conchillo Flaqué
dc7bf98ce5 Pipeline: improve performance by using direct mode 2025-08-20 07:40:21 -07:00
Aleix Conchillo Flaqué
609a43a191 FrameProcessor: added processors/next/previous properties 2025-08-20 07:40:19 -07:00
Aleix Conchillo Flaqué
4fb04422d9 FrameProcessor: remove unused set_parent/get_parent 2025-08-20 07:40:02 -07:00
Mark Backman
5205f56087 Add apply_text_normalization to ElevenLabs TTS services 2025-08-19 21:19:00 -04:00
Mark Backman
694c792af3 Merge pull request #2470 from pipecat-ai/mb/11labs-settings-reconnect
Update ElevenLabsTTSService: update runtime configuration
2025-08-19 18:18:14 -07:00
Mark Backman
837de5f893 Merge pull request #2468 from pipecat-ai/mb/fix-mistral-docs-errors
Fix Mistral docstrings build errors
2025-08-19 12:22:26 -07:00
Mark Backman
b3e4421191 Add retry_on_timeout to AWSBedrockLLMService 2025-08-19 14:20:35 -04:00
Mark Backman
84058c3948 Add retry_on_timeout to AnthropicLLMService 2025-08-19 14:20:35 -04:00
Mark Backman
aebc781419 Update ElevenLabsTTSService to update when voice_settings change 2025-08-19 13:51:10 -04:00
Mark Backman
4160446f4c Update ElevenLabsTTSService: reconnect on model and language changes 2025-08-19 11:32:54 -04:00
Mark Backman
05a14af184 Fix Mistral docstrings build errors 2025-08-19 10:31:03 -04:00
Filipi Fuchter
f550015efb Changing heygen log level to trace. 2025-08-18 18:00:25 -03:00
vipyne
088cb56922 log: add Disconnected from ElevenLabs debug log 2025-08-15 15:05:07 -05:00
Aleix Conchillo Flaqué
a789e5feea Merge pull request #2451 from pipecat-ai/aleix/audio-buffer-processor-overlap
AudioBufferProcessor: fix overlap when buffer size is set
2025-08-14 15:31:50 -07:00
Aleix Conchillo Flaqué
16ca44131c Merge pull request #2452 from pipecat-ai/aleix/runner-daily-direct-handlesigint
Runner: set handle_sigint to True for Daily direct
2025-08-14 15:25:05 -07:00
Aleix Conchillo Flaqué
e2fc8b3dce Runner: set handle_sigint to True for Daily direct 2025-08-14 14:55:52 -07:00
Aleix Conchillo Flaqué
8b641089f8 AudioBufferProcessor: fix overlap when buffer size is set 2025-08-14 14:44:08 -07:00
Mark Backman
7f0494aa04 Override build_chat_completion_params for Mistral 2025-08-14 10:32:18 -04:00
Mark Backman
2b2b0f8121 Add MistralLLMService 2025-08-14 09:57:14 -04:00
Ashot
3b55d16a49 fix(asyncai): force flush WS TTS to eliminate stalls 2025-08-14 16:34:34 +04:00
Mark Backman
d7f31e0cbd Merge pull request #2387 from pipecat-ai/mb/retry-chat-completion
Retry chat completions for OpenAILLMService and its subclasses
2025-08-13 14:39:40 -07:00
Mark Backman
71cd0f1c87 fix: Add text support to OpenAIRealtimeBetaLLMService 2025-08-13 11:37:36 -04:00
Filipi Fuchter
bbbbdc459a Fixed an issue where SmallWebRTCTransport ended before TTS finished. 2025-08-13 11:46:51 -03:00
Mark Backman
d203528dad Merge pull request #2333 from yohan-altrium/fix/2277-azure-tts-ssml-reserved-characters
Fixes 2277 - SSML reserved characters causes Azure TTS to fail
2025-08-13 06:27:30 -07:00
Yohan Liyanage
4bcca7956e Refactors the code based on PR comments and adds the relevant changelog entry. 2025-08-13 16:34:33 +05:30
Aleix Conchillo Flaqué
0508ddddfb WatchdogPriorityQueue: fix watchdog sentinel insertion
We now force each inserted item in the priority queue to be a tuple and the
actual value to be last in the tuple. All the previous values in the tuple also
need to be numeric.
2025-08-12 17:40:58 -07:00
Mark Backman
8714c9137f Code review fixes 2025-08-12 17:49:13 -04:00
Mark Backman
4c029fcfa7 Update OpenAILLMService subclasses to use the new build_chat_completion_params function 2025-08-12 17:48:51 -04:00
Mark Backman
5c86f8e687 Add timeout/retry logic and refactor parameter building in BaseOpenAILLMService
- Add timeout (default 5.0s) and retry_on_timeout parameters to constructor
- Implement timeout/retry logic in get_chat_completions using asyncio.wait_for
- Extract build_chat_completion_params() as public method for subclass customization
2025-08-12 17:48:51 -04:00
Mark Backman
54a4d8a9f8 Merge pull request #2422 from thsunkid/thu/fix-set-lang-in-base-whisper
Fix: assigns string code instead of Language enum to BaseWhisperSTTService._language
2025-08-12 11:57:46 -07:00
Mark Backman
38af514d95 Merge pull request #2407 from pipecat-ai/mb/add-gemini-tts
Add GeminiTTSService
2025-08-12 11:56:45 -07:00
Aleix Conchillo Flaqué
6aa80c0b8e Merge pull request #2424 from pipecat-ai/aleix/system-frame-queues-fix
FrameProcessor: fix race condition on FrameProcessorQueue
2025-08-12 11:56:00 -07:00