Skip to content

Trace end-of-utterance waits in LiveKit Agents and Pipecat - #1145

Open
David Elner (delner) wants to merge 3 commits into
mainfrom
delner/audio-eou-wait
Open

David Elner (delner) wants to merge 3 commits into
mainfrom
delner/audio-eou-wait

Conversation

@delner

Copy link
Copy Markdown
Contributor

Add eou_wait task spans to show time spent deciding whether a user has finished speaking, separately from transcription and model execution.

Each pause produces one wait under its user turn. Resumed speech closes the wait; a later pause starts another:

user_turn                            1.0–4.0
  user_speech                        1.0–2.0
  eou_wait outcome=user_resumed       2.0–2.3
  user_speech                        2.3–3.5
  eou_wait outcome=committed          3.5–4.0

Spans use framework-reported speech timing without changing turn-taking behavior. Unsupported timing or outcome metadata is omitted.

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Nov 18, 2026 •

Copy link
Copy Markdown
Codex Review Summary: ✅ Completed

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-11-18T23:45:32.578280Z 18e8536 PR opened
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant