Get a capture
Returns the current document for one capture. Unlike the list — which is a snapshot stream — this always reflects current state, so if a speaker is renamed after the fact, re-fetching self-heals. Only captures in your workspace’s egress stream are fetchable.
Authorizations
Workspace API key (parable_...), minted in the Parable app under Settings → API.
Path Parameters
Capture id (UUID).
Response
The transcript document.
One finished capture with its full transcript.
Capture id — the stable key for deduplication.
ambient = device-initiated on speech; manual = user-initiated recording.
ambient, manual Model-generated after processing; falls back to a generic label.
Model-generated; may be null.
The capturing device — the room, in practice: devices are named after where they live.
The source of truth — time-ordered utterances with speaker attribution.
Convenience plain-text rendering of the segments, one speaker-labeled line per consecutive same-speaker run.
Reserved for the Google Calendar integration; always null today. Additive later — no schema break.
When this document was rendered.