Implement intelligent interruption handling logic#492
Open
SakashSrivastava wants to merge 1 commit intoDark-Sys-Jenkins:mainfrom
Open
Implement intelligent interruption handling logic#492SakashSrivastava wants to merge 1 commit intoDark-Sys-Jenkins:mainfrom
SakashSrivastava wants to merge 1 commit intoDark-Sys-Jenkins:mainfrom
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Implemented intelligent interruption handling that distinguishes between passive acknowledgements (filler words) and active interruptions based on agent speaking state.
Changes Made
New Files:
Modified Files:
livekit-agents/livekit/agents/voice/agent_activity.py`**:
Added early checks in
on_vad_inference_doneto prevent audio breaksUltra-early filler word detection in
_interrupt_by_audio_activityPrevents calling interruption logic when transcript is empty or contains filler words
Added debouncing to prevent duplicate processing
examples/voice_agents/basic_agent.py`**:
Updated to use Deepgram STT/TTS and Groq LLM
Configured for testing interruption scenarios
livekit-agents/livekit/agents/ipc/supervised_proc.py`**:
Fixed Windows signal handling issue
Features Implemented
✅ Scenario 1: Agent ignores "yeah/ok/hmm" while speaking (no audio breaks, pauses, or stutters)
✅ Scenario 2: Agent responds to "yeah" when silent (treats as valid input)
✅ Scenario 3: Agent interrupts immediately for commands like "stop" or "no"
✅ Scenario 4: Agent handles mixed inputs like "yeah okay but wait" (interrupts due to "wait")
Configuration
The ignore list is configurable via environment variable:
export AGENT_IGNORE_WORDS="yeah,ok,hmm,right,uh-huh,aha,yep,yup,okay"
Proof
A short video demonstrating all required test scenarios was submitted via the official Google Form.