Skip to main content

Run Transcription

This endpoint is used to transcribe an audio file with optional alignment and speaker diarization.

Request​

URL:
POST /transcription

Headers:
Authorization: Bearer <TOKEN>

Request Body Parameters​

The request must include a JSON body with the following parameters:

  • audio_path (required): The URL of the audio to be transcribed. URL can be public video URL.
  • task (optional, default: "transcribe"): Specifies the task to perform. The default is "transcribe".
  • language (optional, default: "en"): Defines the language of the transcription.
  • initial_prompt (optional, default: None): An initial text prompt to guide the transcription process.
  • align (optional, default: true): If set to true, enables word alignment in the transcript.
  • diarize (optional, default: true): If set to true, enables speaker diarization, identifying different speakers in the transcript.

Example Request​

curl --location --request POST 'https://platform.recordly.ai/transcription' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer <TOKEN>' \
--data-raw '{
  "audio_path": "<AUDIO_PATH>",
  "task": "transcribe",
  "language": "<LANGUAGE>",
  "initial_prompt": None,
  "align": true,
  "diarize": true
}'

Example Response​

If the request is successful, the response includes a job_id and a result_url where the transcription can be retrieved.

{
  "job_id": "daf413d1-be10-4467-8076-f4f59d48aad5",
  "result_url": "https://platform.recordly.ai/transcription/daf413d1-be10-4467-8076-f4f59d48aad5"
}