YouTube Transcript Scraper — LLM-Ready Video Text

Turn any YouTube video into clean, structured text you can drop straight into ChatGPT, Claude, or your own RAG pipeline. Give it video links, a whole playlist, a channel, or a search query — get back the full transcript, timestamps, and metadata.

Built for the thing millions of people do every day: "summarize this video," "pull the key points," "search across these 40 talks."

What you get per video

Field	Description
transcript	Clean, joined plain-text transcript — ready to paste into an LLM
segments	Timestamped caption lines (start, duration, text)
markdown	Ready-to-paste Markdown (title + channel + link + transcript)
language / isGenerated	Transcript language and whether it's auto-generated
title, channel, channelUrl	Video and creator info
durationSeconds, viewCount, publishDate	Video metadata
description, thumbnail, wordCount	Extra context

Inputs

• Video URLs or IDs — watch?v=, youtu.be/, Shorts, embed, or a bare 11-char ID.
• Playlist URLs — expanded into their videos.
• Channel URLs — recent uploads from @handle or /channel/UC....
• Search queries — top matching videos for each query.
• Preferred languages + optional auto-translation.
• Timestamps / Markdown toggles.
• Proxy — residential recommended (YouTube blocks transcript requests from datacenter IPs).

Use it with your LLM

The dataset is JSON, so you can pipe results straight into any model. Example — get results via the Apify API and hand the markdown field to your LLM:

$curl "https://api.apify.com/v2/datasets/<DATASET_ID>/items?format=json"

Every item's transcript and markdown fields are pre-cleaned for token efficiency (HTML entities decoded, newlines collapsed). Perfect for summarization, Q&A over videos, content repurposing, and building searchable knowledge bases.

Pricing

Pay-per-result: you're charged once per successfully extracted transcript (transcript-scraped event). Videos with no captions available are not charged. Set the per-result price in the Apify Console monetization settings — a low price (e.g. a fraction of a cent per transcript) keeps it affordable for daily, high-volume use.

Frequently asked questions

How do I get a transcript from a YouTube video? Paste one or more video URLs (or IDs) and the Actor returns the full transcript as clean text and Markdown, ready for your LLM.

Do I need a YouTube API key or login? No. No API key, no login, no cookies — it reads publicly available captions.

Does it work with auto-generated captions? Yes — both manually-created and auto-generated captions, with fallback to any available language.

Can I get YouTube transcripts as JSON, CSV or plain text? Yes. Results are a dataset you can export to JSON, CSV, Excel, or pull via the API — the transcript and markdown fields are pre-cleaned for LLMs.

Can I transcribe many videos at once? Yes — pass a list of URLs (or a playlist/channel's video URLs) for bulk extraction; you're only charged for videos that have captions.

What can I use the transcripts for? Summarization, Q&A over videos, RAG knowledge bases, content repurposing, subtitles and research.

Notes

• Works with both manually-created and auto-generated captions.
• Falls back to any available language when your preferred languages aren't present.
• Metadata is best-effort and never blocks a transcript from being returned.
• Respect YouTube's Terms of Service and creators' rights when using scraped content.

⭐ Found this useful?

If this Actor saved you time, please leave a rating on its Apify page — reviews genuinely help others discover it and help me keep improving it.

🤖 Use with Claude or ChatGPT (MCP)

Run this actor from Claude, ChatGPT, Cursor or any MCP client via the Apify MCP server. In Claude Desktop: Settings → Connectors → Add custom connector → https://mcp.apify.com. Or expose just this tool:

{ "mcpServers": { "apify": { "url": "https://mcp.apify.com?tools=get_anything/youtube-transcript-scraper" } } }

Full guide: Connect Apify actors to Claude & ChatGPT.