POST
/
api
/
v1
/
client
/
transcribe
Unified Transcription
curl --request POST \
  --url http://localhost:8000/api/v1/client/transcribe \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: multipart/form-data' \
  --form include_ts=true \
  --form model=WhisperLargeV3 \
  --form 'source_url=<string>' \
  --form source_file='@example-file' \
  --form return_result_in_response=true \
  --form 'webhook_url=<string>'
{
  "data": {
    "request_id": "550e8400-e29b-41d4-a716-446655440000"
  }
}

Authorizations

Authorization
string
header
required

API key obtained from the ModelBeam dashboard

Body

multipart/form-data
include_ts
boolean
required

Include timestamps in transcription

model
string
required
Example:

"WhisperLargeV3"

source_url
string

URL to transcribe (YouTube, X/Twitter, Twitch, Kick, TikTok, X Spaces). Mutually exclusive with source_file

source_file
file

Audio or video file to transcribe. Mutually exclusive with source_url

return_result_in_response
boolean

Return transcription directly in the response

webhook_url
string
Maximum string length: 2048

Response

Transcription job created

data
object