YouTube Transcript Scraper: videos and whole channels, text, segments and SRT

Get the transcript of any YouTube video, or of a channel's latest videos, as clean text, timed segments or a ready-made SRT subtitle file. You choose your preferred languages, and can use YouTube's auto-captions or its built-in translation.

There's no Actor fee: you only pay Apify platform usage. In our test that was about $0.45 per 1,000 transcripts, residential proxy included. You also don't need an API key.

What you get

• 📝 The full transcript as one clean string, ready for summarization, RAG or embeddings
• ⏱️ Timed segments (start, duration, text) and an optional SRT subtitle file
• 🌍 Language preference: manual captions first, then auto-generated, then any available language (each step is configurable)
• 🔁 Translation of the captions into any language, using YouTube's own machine translation
• 📺 Whole channels: give a @handle and get transcripts of its latest videos
• 📊 Video details with every transcript: title, channel, length, exact view count and every caption language available
• ✅ A clear status for every video: OK, NO_TRANSCRIPT, NO_MATCHING_LANGUAGE, AGE_RESTRICTED, VIDEO_UNAVAILABLE, LIVE_STREAM, EMPTY_TRANSCRIPT or FAILED

Tested at scale

40 transcripts out of 40 videos came back in 106 seconds, from single videos plus two whole channels. That used $0.018 of platform usage.

Use cases

• AI and RAG: turn a channel into a searchable knowledge base.
• Content repurposing: turn videos into blog posts, newsletters, threads or chapters.
• Research and monitoring: analyze what creators, competitors or experts actually say.
• Subtitles: download SRT files, optionally translated.

Input

{
  "videoUrls": ["https://www.youtube.com/watch?v=aircAruvnKk", "jNQXAC9IVRw"],
  "channels": ["@kurzgesagt"],
  "maxVideosPerChannel": 20,
  "preferredLanguages": ["en", "pt"],
  "allowAutoGenerated": true,
  "fallbackToAnyLanguage": true,
  "translateTo": "",
  "includeSegments": true,
  "includeSrt": false
}

Output

{
  "videoId": "aircAruvnKk",
  "title": "But what is a neural network? | Deep learning chapter 1",
  "channelName": "3Blue1Brown",
  "lengthSeconds": 1120,
  "viewCount": 25097073,
  "status": "OK",
  "languageCode": "en",
  "isAutoGenerated": false,
  "isTranslated": false,
  "wordCount": 3357,
  "text": "This is a 3. It's sloppily written and rendered at an extremely low resolution of 28x28...",
  "segments": [
    {
      "start": 4.22,
      "duration": 1.18,
      "text": "This is a 3."
    }
  ],
  "availableLanguages": [
    {
      "languageCode": "ar",
      "language": "Arabic",
      "isAutoGenerated": false
    },
    {
      "languageCode": "bn",
      "language": "Bangla",
      "isAutoGenerated": false
    }
  ]
}

Why residential proxy?

YouTube shows a "confirm you're not a bot" check to caption requests from cloud and datacenter IPs. That's why the Actor uses Apify residential proxy by default and rotates IPs automatically. Responses are small, so the proxy costs fractions of a cent per video.

Pricing

This Actor has no fee of its own: you pay only Apify platform usage, meaning compute and residential proxy. In our test, 40 transcripts cost $0.018.

More YouTube and trends tools from Criato Labs

• YouTube Channel & Video Scraper: channel profiles, videos, Shorts and keyword search
• YouTube Comments Scraper: comments and replies
• Google Trends Batch Scraper: including YouTube Search trends

FAQ

What if a video has no captions? You'll get a row with status: NO_TRANSCRIPT. Videos without speech, or whose creator turned off auto-captions, have none.

Does it transcribe audio itself (Whisper)? No. It returns YouTube's own captions, manual or auto-generated, which covers the vast majority of spoken videos.

Can I get transcripts in another language? Yes. Set translateTo (for example en) to use YouTube's machine translation of the original captions.