Features

Supported sources

Accepted URLs, IDs, handles, and queries for every TranscriptFetch endpoint.

Source coverage

Each platform card shows the input forms the API recognizes and which listing endpoints are available.

Choose the endpoint by intent

Use the endpoint for the resource you want back. Each route accepts the complete URL or identifier shown below.

You do not need to extract IDs yourself. Full URLs are parsed server-side. A YouTube watch URL containing list= can be sent to /api/v2/transcripts/video for that single video or /api/v2/transcripts/playlist for the full playlist.

Best-effort video pages

The supported transcript sources are YouTube, TikTok, Instagram, and Direct media file.

Other public HTTP or HTTPS video pages are attempted with the generic extractor. This is best-effort compatibility, not a guarantee that every website or embedded player will work.

Where AI transcription applies

When a supported source has no captions, the AI transcription fallback covers YouTube, TikTok, Instagram, and direct media file URLs. Generic video pages remain best-effort extraction targets; AI transcription is not guaranteed for them.