Repository navigation
feat(audio): add shared session recording infrastructure - #1067
David Elner (delner) wants to merge 12 commits into
Conversation
Add positioned PCM recording, middleware, progressive v3 export, bounded native workers and direct attachment uploads. Include PCM/WAV support, cancellation and upload lifecycle coverage, and audio-only attachment request timeouts.
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour 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. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 6c033ff19a
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "Codex (@codex) review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "Codex (@codex) address that feedback".
- clean up some export logic - reduce repeated middleware processing
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: a67658b923
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "Codex (@codex) review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "Codex (@codex) address that feedback".
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 14aa263c5f
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "Codex (@codex) review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "Codex (@codex) address that feedback".
Adds a common recording layer that turns caller and agent audio into an aligned
session recording, exported progressively as attachments. This is the foundation
for the LiveKit integration and future Pipecat recording work.
An adapter supplies audio, its position in the conversation, and a capture policy:
The session accumulates timed recording segments with aligned audio channels:
Design
flowchart LR P[Positioned PCM] --> R[Shared timeline / recorder] R --> W[Background normalization / encoding] W --> U[Attachment upload] U --> S[Session references + segment metadata]AudioSessionRecordermaintains a timeline that is used to preserve silence, channel alignment and source identity across audio streams with different sample rates.