YouTube Transcript

Turn a public YouTube video URL or ID into its full transcript: timestamped caption segments plus a single joined text block, with the language and whether the track is auto-generated.

Socialv0.1.2~1 credit/runSource on GitHub

Overview

YouTube Transcript turns any public YouTube video into clean, structured text. It reads the video's caption tracks through the Better Fetch stealth engine, selects the best track for your language, and returns both timestamped segments and one joined transcript string — the ideal input for summarization, search indexing, retrieval-augmented generation, or content repurposing without downloading media.

Last validated: Jul 3, 2026

Playground

Input

urlstring

Public YouTube video URL (watch, youtu.be, shorts, or embed). Use url or video_id.

langstring

Preferred caption language code, e.g. en or es. Falls back sensibly when unavailable.

video_idstring

11-character YouTube video id, as an alternative to url.

max_segmentsintegerdefault: 3000

Maximum caption segments to return. The joined text field always covers the returned segments.

Output

urlstringrequired

Canonical watch URL used

langstring

Language code of the chosen caption track

textstringrequired

All segments joined into one transcript string

titlestring

Video title

authorstring

Channel name

segmentsobject[]required

Timestamped caption cues

video_idstringrequired

YouTube video id

is_generatedboolean

True if the track is auto-generated (ASR)

segment_countintegerrequired

Number of caption segments returned

duration_secondsnumber

Video length in seconds

Examples

transcript-by-url

{
  "url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ"
}

Use cases

Summarize long videos

Pull a full transcript and feed it to an LLM to produce summaries, chapter notes, or key takeaways without watching the video.

Search and RAG indexing

Index timestamped segments so agents can retrieve and cite the exact moment a topic is discussed inside a video.

Content repurposing

Convert talks, tutorials, and interviews into text for blog drafts, show notes, and quote extraction.

FAQ

Does it work on any YouTube video?

It works on public videos that have captions — either human-written or YouTube's auto-generated track. Videos with captions disabled, or private and age-restricted videos, cannot be transcribed.

Can I choose the transcript language?

Yes. Pass a language code such as en or es and the tool prefers a matching track, falling back to a manual English track, then any manual track, then auto-generated captions.

What is the difference between the segments and the text field?

segments are timestamped caption cues with start and duration in seconds, useful for citation and search. text is every segment joined into one string, convenient for summarization prompts.

Use it anywhere

MCP (Claude, Cursor, any client)

# Add the Better Fetch MCP connector (or paste the URL into
# Claude → Settings → Connectors → Add custom connector):
claude mcp add --transport http better-fetch https://betterfetch.co/api/mcp \
  --header "Authorization: Bearer bf_your_key_here"

# Then ask for the tool by name: youtube_transcript

REST

curl -sS -X POST "https://betterfetch.co/api/tools/youtube_transcript/run" \
  -H "Authorization: Bearer bf_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{"input": {"url":"https://www.youtube.com/watch?v=dQw4w9WgXcQ"}}'

Run locally

git clone https://github.com/better-fetch/tools/tree/main/tools/youtube-transcript && cd youtube-transcript && npm i
BETTER_FETCH_API_KEY=bf_your_key_here npx bf-tool run --input '{"url":"https://www.youtube.com/watch?v=dQw4w9WgXcQ"}'