API resource
Transcription jobs
Queue audio, video, public links, or completed uploads through one asynchronous job model.
One lifecycle for every supported source
A URL job and an uploaded-file job produce the same status and result shape. Your application can keep one durable job record while Fast Transcriber handles source resolution, media processing, and speech recognition.
Two source modes
Use a public source_url or an account-owned upload reference—never both.
Asynchronous by design
Creation returns 202 immediately so long recordings do not hold an HTTP request open.
Consistent results
Completed jobs expose full text, duration, and timestamped segments in one JSON envelope.
Account scoped
Jobs and upload keys are isolated to the account represented by the bearer key.
From request to transcript
A predictable sequence you can reuse in server applications, automations, and internal tools.
- 01
Choose the source
Use source_url for public media or prepare and finish a direct file upload.
- 02
Create the job
POST exactly one source to /api/v1/transcriptions.
- 03
Poll to a terminal state
GET the Location URL until status is completed or failed.
Make the request
Examples use the production API origin and an environment variable for the secret key.
curl --include --request POST https://fast-transcriber.com/api/v1/transcriptions \
--header "Authorization: Bearer $FAST_TRANSCRIBER_API_TOKEN" \
--header "Content-Type: application/json" \
--data '{
"source_url": "https://www.youtube.com/watch?v=VIDEO_ID",
"speaker_diarization": false
}'Requirements and boundaries
- Send an active API key in the Authorization bearer header.
- Provide exactly one of source_url or upload.
- Respect the authenticated account's daily use, file-size, and feature limits.
- Treat 202 as queued—not completed—and persist the returned job ID.
Frequently asked questions
Does job creation return the transcript?+
No. It returns 202 Accepted with a processing job. Retrieve that job until it completes.
Can one request contain several files?+
No. The current v1 contract creates one job from one URL or one uploaded object.
Can I request speaker identifiers?+
Yes on Pro by setting speaker_diarization to true when creating the job.