Introduction

Transcript turns a YouTube, X, TikTok, Instagram, RedNote, or Facebook link into a clean, timestamped transcript — in seconds, in any language. Captions are used when a platform provides them; otherwise the audio is transcribed with ASR automatically, no extra step.

Ways to use it

Web app

Paste a link, get a transcript. No install.

CLI

transcript get <url> from your terminal — great for scripting and piping.

REST API

POST /v1/transcript — call it from any language.

Quickstart

Get your first transcript in under a minute.

How it works

Send a URL. POST /v1/transcript with a platform link (or use the CLI/web app, which call the same endpoint).
Get a transcript, fast. If captions exist, they come back immediately. If not, the API queues an ASR job and returns 202 with a job_id — poll GET /v1/jobs/{job_id} until it's done.
Pick a format. json, txt, srt, or vtt.

Authentication

Every request is either:

Signed in on the web app (a Supabase session), or
A product API key (sk_live_...), sent as Authorization: Bearer sk_live_....

Both resolve to the same account and the same monthly transcription budget. The CLI's transcript login mints/reuses that same sk_live_... key via a browser device-login flow — see CLI: Authentication.

An unauthenticated request still works, capped to a short preview — useful for trying the API before signing up.