fix(voice): clear forwardedText when interrupted before any text syncs#1270
Merged
fix(voice): clear forwardedText when interrupted before any text syncs#1270
Conversation
When the first audio frame was queued but interruption hit before the synchronizer produced any text, forwardedText retained the full unplayed LLM response and was committed to chat context — causing the agent to 'remember' things it never said.
Contributor
Author
|
closes #1269 |
🦋 Changeset detectedLatest commit: 3eb5e1a The changes in this PR will be included in the next version bump. This PR includes changesets to release 25 packages
Not sure what this means? Click here to learn what changesets are. Click here if you're a maintainer who wants to add another changeset to this PR |
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
theomonnom
approved these changes
Apr 17, 2026
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #1269
Description
When the user interrupts the agent immediately after the first audio frame is queued but before the transcription synchronizer has produced any text, the full unplayed LLM response was incorrectly inserted into the chat context. This caused the agent to "remember" things it never actually said.
The fix clears
forwardedTextin theelsebranch whenplaybackEv.synchronizedTranscriptis empty, rather than letting it retain the initial value oftextOut?.text(the full generated response).Changes Made
agent_activity.ts(standard pipeline): clearforwardedText = ''whensynchronizedTranscriptis empty after first frame was playedagent_activity.ts(realtime path): same fix for the realtime generation branchPre-Review Checklist
Testing
Additional Notes
The Python
livekit-agentsrepo has the same structural bug (if synchronized_transcript is not None:without anelse) — worth mirroring upstream.Note to reviewers: Please ensure the pre-review checklist is completed before starting your review.