Skip to content

Add audio traces - #88

Merged
David Elner (delner) merged 1 commit into
mainfrom
delner/audio_traces
Nov 18, 2026
Merged

David Elner (delner) merged 1 commit into
mainfrom
delner/audio_traces

Conversation

@delner

Copy link
Copy Markdown
Contributor

No description provided.

@delner

Copy link
Copy Markdown
Contributor Author

Current version of this PR is a rough, initial draft: a port of my v3 audio trace schema, which was designed primarily to be a human-readable document. It describes:

  • Support for cascade, half-cascade, realtime modes which should be considered stable.
  • Support for duplex mode which is unstable (currently in development, changes expected)
  • Spans that should be created by audio agent frameworks.
    • ...and how audio recordings, timings, relationships, framework-specific data should be attached to these spans.

It does not describe llm spans that are produced at the LLM provider level (e.g. OpenAI SDK). That should come from the existing instrumentation spec for now.

Looking for guidance on direction, how this best fits into the conventions of this repo.

@delner
David Elner (delner) force-pushed the delner/audio_traces branch 2 times, most recently from 5e7f511 to 26b5d92 Compare November 18, 2026 03:47
@delner
David Elner (delner) merged commit 93867ec into main Nov 18, 2026
4 checks passed
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.

2 participants