Issues

Runtime-added observers receive no frame events when the observer set is empty

#5860

· lkrr1213 opened

on Sep 20, 2026

Aggregator triggers inference before all function calls in a batch have started when the first handler never yields

#5852

· dp90 opened

on Sep 19, 2026

RNNoiseFilter: Graph.init() got an unexpected keyword argument 'rate' on macOS Python 3.12

Bug

#5849

· gear12beer opened

on Sep 18, 2026

Feature request: expose provider response/request ids (OpenAI Responses, Deepgram, ElevenLabs) without DEBUG logging

#5843

· mannyb223 opened

on Sep 18, 2026

FunctionCallResultFrame arriving while the user is still speaking is silently dropped, the LLM is never re-invoked and nothing retries

Bug

#5823

· SwAt1563 opened

on Sep 17, 2026

Feature request: safe opt-in empty-user-turn recovery via a developer message and one LLM run

#5819

· timofey-TK opened

on Sep 17, 2026

Twilio/Telnyx/Exotel serializers crash the whole call on a malformed WebSocket message

Bug

#5816

· wankhede04 opened

on Sep 17, 2026

Is there a documented recovery when an LLM stream starts but stalls before a sentence completes?

#5804

· mannyb223 opened

on Sep 16, 2026

Is PipelineWorker.remove_observer() supported for an observer to detach itself from inside its own event handler?

#5803

· mannyb223 opened

on Sep 16, 2026

DeepgramFluxSTTService: a blocked websocket send (peer stops reading, socket in CLOSE_WAIT) stalls the whole pipeline with no error, no reconnect and no log line

#5796

· mannyb223 opened

on Sep 16, 2026

No public way to settle outstanding function calls before `on_pipeline_finished` runs

#5790

· mannyb223 opened

on Sep 16, 2026

Telnyx serializer's interruption write has no timeout: a blocked socket wedges the pipeline until the peer closes TCP, and cancel does not reach it

#5789

· mannyb223 opened

on Sep 16, 2026

Navigation Menu

All issues

Runtime-added observers receive no frame events when the observer set is empty

Aggregator triggers inference before all function calls in a batch have started when the first handler never yields

RNNoiseFilter: Graph.init() got an unexpected keyword argument 'rate' on macOS Python 3.12

Feature request: expose provider response/request ids (OpenAI Responses, Deepgram, ElevenLabs) without DEBUG logging

FunctionCallResultFrame arriving while the user is still speaking is silently dropped, the LLM is never re-invoked and nothing retries

Feature request: safe opt-in empty-user-turn recovery via a developer message and one LLM run

Twilio/Telnyx/Exotel serializers crash the whole call on a malformed WebSocket message

Is there a documented recovery when an LLM stream starts but stalls before a sentence completes?

Is PipelineWorker.remove_observer() supported for an observer to detach itself from inside its own event handler?

DeepgramFluxSTTService: a blocked websocket send (peer stops reading, socket in CLOSE_WAIT) stalls the whole pipeline with no error, no reconnect and no log line

No public way to settle outstanding function calls before `on_pipeline_finished` runs

Telnyx serializer's interruption write has no timeout: a blocked socket wedges the pipeline until the peer closes TCP, and cancel does not reach it

Search results