Skip to main content

`POST /api/v1/stt/transcribe`, transcribe an uploaded audio file.

POST 
/api/v1/stt/transcribe

Accepts multipart/form-data with:

  • audio (required): WAV audio file
  • language (optional): language hint (ISO 639-1 code)
  • source (optional): what the audio is, api (default), file or hotkey, the tag the history shows; the desktop attached to this runtime sends a file it read or a dictation it captured

Returns 200 OK with the persisted transcript row (TranscriptRow). Returns 400 Bad Request on missing or invalid audio, or an unknown source. Returns 503 Service Unavailable when the engine is absent.

Request​

Responses​

Persisted transcription row