Get a YouTube video transcript
Full transcript for one video, plus title/author/thumbnail metadata. One block of text by default; add 'timestamps=true' to also get 'segments', one {start, duration, text} per caption line (times in seconds). Same credit cost either way. Provenance: language_code is the caption track actually returned and can differ from requested_language (YouTube falls back to the track it has); caption_type is 'manual' (creator-uploaded) or 'auto' (speech recognition), null when unknown for older transcripts; cached says whether it came from our stored copy, fetched at fetched_at. Failed requests (no captions, unavailable video) are never charged.
Request
- Base URL: https://getyoutubetranscript.com/api/v1
- URL: https://getyoutubetranscript.com/api/v1/transcript
- Auth: one of:
- HTTP bearer
- API key in header x-api-key
Query parameters
YouTube video URL (full or short) or an 11-character video ID.
Caption language code, e.g. 'en', 'es'.
Set to 'true' to include 'segments': per-line {start, duration, text} timestamps in seconds, e.g. for citing a moment in the video. Off by default because it roughly doubles the response size.
Language code to machine-translate the transcript into, e.g. 'es', 'de', 'zh-CN'. Each caption line is translated in place, so timestamps stay aligned. The response adds 'translated_to'; 'language_code' stays the source track. Same 1-credit cost.
Response
Transcript retrieved. 'segments' is present only when timestamps=true.
Changes
No recorded changes to this operation across all 1 revision of this API.