YouTube Transcript Scraper: captions as text or timestamped lines

Paste YouTube links, get the transcripts back. One row per video with the full text, or one row per caption line with its start time if you need timestamps.

You pay once per transcript delivered. Videos with no captions, private videos and bad links cost nothing.

What you get

Field	Example
title	Me at the zoo
channelName	jawed
transcript	All right, so here we are, in front of the elephants...
wordCount	39
language	en
isAutoGenerated	false
availableLanguages	en, de
durationSeconds / viewCount	19 / 443223613
videoUrl	Link to the video

Pick timestamped lines and each row also carries segmentIndex, startSeconds, durationSecondsSegment and text, which is what you want for subtitles, clip finding or chaptering.

Sample output (full text mode)

{
  "videoId": "jNQXAC9IVRw",
  "videoUrl": "https://www.youtube.com/watch?v=jNQXAC9IVRw",
  "title": "Me at the zoo",
  "channelName": "jawed",
  "channelId": "UC4QobU6STFB0P71PMvOGN5A",
  "durationSeconds": 19,
  "viewCount": 443223613,
  "language": "en",
  "isAutoGenerated": false,
  "availableLanguages": "en, de",
  "transcript": "All right, so here we are, in front of the elephants the cool thing about these guys is that they have really... really really long trunks and that's cool...",
  "wordCount": 39
}

Quick start

1. Paste video links or IDs into Videos, one per line. Watch links, youtu.be, Shorts and live links all work.
2. Set Preferred language if you want something other than English.
3. Click Start.

The default input runs as-is.

Use cases

• Turn your own back catalog into blog posts, show notes or captions.
• Search a set of talks or interviews for every mention of a product or name.
• Build subtitle files from the timestamped lines.
• Pull quotes with exact timestamps for research, reporting or citation.
• Check what was said in a video without watching all of it.

Languages and auto captions

Most videos only have YouTube's automatic captions, and those are on by default. isAutoGenerated tells you which kind you got. Turn Allow auto-generated captions off if you only want human-made ones.

If your language isn't on the video, you get the closest track and the language field says which. availableLanguages lists every track so you can re-run in another one.

What happens when something goes wrong

Each failed video is listed in RUN_SUMMARY (in the run's key-value store) with the reason: no captions, private or removed, or not a YouTube link. You're not charged for any of them, and the rest of the batch keeps going.

YouTube sometimes asks servers to "confirm you're not a bot". When that happens the scraper switches to a different connection and retries on its own. In our tests that took 20 of 20 videos through.

Pricing

One charge per transcript delivered, plus Apify's small start fee per run. Timestamped mode is still one charge per video, however many lines it has. Set a maximum cost per run and the scraper stops at your limit.

FAQ

Does it need a YouTube account or API key? No.

Can it do a whole channel or playlist? Not yet. Paste the video links. Channel and playlist input is next on the list.

How accurate are auto captions? As accurate as YouTube's. Music, heavy accents and crosstalk make them worse. We don't edit the text.

Something broke? Open an issue with the video link. I answer within a day.

Limitations

• Videos with captions turned off have no transcript to fetch.
• Age-restricted and members-only videos are skipped.
• Very long videos (several hours) still work, but take a few seconds each.

Export or use as an API

Download any run's results from the Storage tab as CSV, Excel, JSON or XML.

To call it from your own code, send your input to Apify's API and get the rows back in one request:

POST https://api.apify.com/v2/acts/leathered_latch~youtube-transcripts-scraper/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN

It also plugs into Make, Zapier and n8n through Apify's integrations, and AI agents can find and run it through Apify's MCP server.

Your responsibility

A transcript is the video creator's work. This tool fetches the captions YouTube already shows publicly; it doesn't give you any rights to them. Quoting, research and your own videos are the usual uses. Republishing someone else's transcript, or using it to train an AI model, may need the creator's permission. That call is yours.

If you own content that shows up in these results and want it handled differently, open an issue on the Issues tab. We answer within 14 days.

Data source and credit

Every row comes from YouTube (Google LLC). We don't own this data, and we don't take credit for it. The captions belong to the creator of each video.

This tool reads only what YouTube shows to any logged-out visitor. No logins, no private data. Each row keeps a link back to the page it came from (the videoUrl field), so you can check any result and credit the source when you use it.

Not affiliated with, endorsed by or sponsored by YouTube or Google. All names and trademarks belong to their owners.

Changelog

• 0.1 First release. Full text or timestamped lines, language choice, auto-caption control, automatic retry through YouTube's bot check.