Referenta Docs | Transcribe a recorded audio clip

Forwards a recorded audio clip to ElevenLabs batch speech-to-text and returns the transcribed text.

Authorization

dashboardSession

sb-access-token<token>

Authenticated dashboard session cookie. Some routes also rely on additional Supabase SSR cookies.

In: cookie

multipart/form-data

TypeScript Definitions

Use the request body type in TypeScript.

audio*file

Recorded audio blob (webm, mp4, or ogg).

Formatbinary

language?string

Optional ISO language code, forwarded as ElevenLabs' language_code.

allowEmpty?string

Sent as the literal string "true" to resolve an empty/silent recording as {"text": ""} instead of a 422 error.

Value in"true" | "false"

curl -X POST "https://example.com/api/v1/assistant/scribe-transcription" \  -F audio="string"

{
  "text": "string"
}

{
  "error": "string",
  "details": "string",
  "receivedKeys": [
    "string"
  ]
}

{
  "error": "string",
  "details": "string",
  "receivedKeys": [
    "string"
  ]
}

{
  "error": "string",
  "details": "string",
  "receivedKeys": [
    "string"
  ]
}

{
  "error": "string",
  "details": "string",
  "receivedKeys": [
    "string"
  ]
}