FastTranscriber

The transcription API for developers and AI agents

Transcribe audio, video, uploaded files, and public links with one asynchronous REST API and predictable structured results.

Get an API key Read the API docs

Building with an AI agent? Open a setup guide and connect over MCP.

Claude Desktop Codex Cursor

Bring media to one transcription workflow

Supported public-link sources use the same transcription endpoint. Local files begin with a secure upload request before being queued.

YouTube transcription API

Public videos and Shorts that open without signing in.

POST/api/v1/transcriptions

Explore this API

Instagram transcription API

Public Reels and video posts available from a shared URL.

POST/api/v1/transcriptions

Explore this API

TikTok transcription API

Public TikTok video links resolved by the same URL endpoint.

POST/api/v1/transcriptions

Explore this API

Facebook video transcription API

Publicly accessible Facebook video links.

POST/api/v1/transcriptions

Explore this API

Google Drive transcription API

Audio or video files shared with anyone who has the link.

POST/api/v1/transcriptions

Explore this API

Direct media transcription API

A public URL that points directly to an audio or video file.

POST/api/v1/transcriptions

Explore this API

File upload transcription API

Prepare a direct or multipart upload, then queue the file.

POST/api/v1/uploads

Explore this API

From source to transcript in three steps

01
Submit the media
Send a public source_url, or upload a file and pass its verified storage reference when you create the job.
02
Track the job
A queued request returns 202 Accepted. Poll the job URL until its status becomes completed or failed.
03
Use the transcript
Read the transcript text and timestamped segments from the completed JSON response in your own product or workflow.

Transcript data that is ready to use

Completed jobs return a predictable JSON object. Work with the whole transcript, precise segments, or both.

text: The complete transcript as plain text.
segments[]: Timestamped start, end, and text values for each segment.
status: The job state, including completed and failed outcomes.
duration_seconds: The detected duration of the processed recording.
id & created_at: Stable job identity and its creation timestamp.
speaker: Speaker identifiers on segments when Pro diarization is requested.

200 OKapplication/json

{
  "data": {
    "id": "95c3fdd8-...",
    "filename": "interview.mp3",
    "status": "completed",
    "duration_seconds": 1842,
    "text": "Completed transcript text...",
    "segments": [
      { "start": 0, "end": 4.2, "text": "Welcome." }
    ]
  }
}

Go from first request to production

Browse all documentation

Quickstart

Create a key and queue your first transcription request.

Authentication

Send your Fast Transcriber key as a bearer token.

API reference

Review endpoints, request bodies, responses, and status codes.

Upload a file

Use direct and multipart uploads for local media.

Errors

Handle the stable API error envelope and failure states.

MCP

Connect Fast Transcriber tools to a compatible AI client.

Build transcription into your next workflow

Create an API key from your Fast Transcriber account, then follow the quickstart to submit your first job.

Get an API key Open the quickstart