YouTube Transcript Scraper — Batch + SRT/VTT Export

Extract YouTube transcripts in bulk — paste a list of video URLs, video IDs, or Shorts links and get back clean transcript text, timestamped segments, and ready-to-use SRT / VTT subtitle files. No YouTube API key, no quota, no browser extension.

Why this YouTube transcript scraper?

• True batch input — transcribe 1 or 500 videos in a single run. Most transcript actors accept a single URL at a time.
• SRT and VTT export — the only formats video editors, subtitle tools, and players actually ingest. Each file is also saved to the run's key-value store for direct download.
• Every URL shape works — watch?v=, youtu.be, Shorts, embeds, live links, or bare 11-character video IDs.
• Failed videos are never charged — private, deleted, or caption-less videos are logged, skipped, and summarized; the batch keeps going.
• Language preference — request en, es, de, or any priority list; manually-created captions are preferred over auto-generated ones, and the result tells you which you got.
• LLM/RAG-friendly output — compact JSON with a clean text field per video, ideal for feeding transcripts into AI agents, summarizers, or vector stores (works out of the box via Apify's MCP server).

Input

{
  "videos": [
    "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
    "https://youtu.be/jNQXAC9IVRw",
    "https://www.youtube.com/shorts/abc123def45"
  ],
  "languages": ["en"],
  "formats": ["text", "srt"]
}

Output

One dataset item per successfully transcribed video:

{
  "videoId": "dQw4w9WgXcQ",
  "url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
  "language": "en",
  "isGenerated": false,
  "text": "We're no strangers to love…",
  "srt": "1\n00:00:18,000 --> 00:00:21,500\nWe're no strangers to love…"
}

Add "segments" to formats for timestamped JSON snippets (text / start / duration), and "vtt" for WebVTT subtitles.

Common use cases

• Subtitle files (SRT/VTT) for video editing, translation, or accessibility workflows
• Bulk transcript datasets for AI summarization, RAG pipelines, and LLM agents
• Content research — search what competitors say across their whole channel
• Repurposing video content into blog posts, newsletters, or show notes

FAQ

Does it need a YouTube API key? No. Transcripts are fetched directly; the official API doesn't expose them anyway.

What about videos without captions? They're reported in the run status message and skipped — you're never charged for a failed video.

Which languages are supported? Anything YouTube has captions for — manual or auto-generated. Pass a preference list like ["de", "en"].

Related actors

• YouTube Outlier Finder — find a channel's breakout videos vs its baseline
• YouTube Channel ID Finder — resolve handles/URLs to canonical channel IDs
• YouTube Featured Channels Scraper — expand a seed channel into its featured channels