YouTube transcripts

Send a YouTube video URL to POST /youtube/transcript. Berrycrawl reads its available captions and returns a timestamped transcript through the standard scrape response shape. The focused endpoint only accepts URLs that identify one YouTube video.

curl https://api.berrycrawl.com/api/v1/youtube/transcript \
  -H "Authorization: Bearer $BERRYCRAWL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://www.youtube.com/watch?v=aircAruvnKk",
    "formats": ["markdown"],
    "timeout": 60000
  }'

The same request is available in the dashboard playground.

Supported URL forms

Berrycrawl recognizes URLs that identify one video.

Channel, playlist, search, and profile URLs do not identify one video and use normal page scraping.

Response

The response uses the same shape as every scrape. Requested text formats appear under data. For Markdown, each caption segment includes its playback timestamp.

{
  "success": true,
  "data": {
    "markdown": "# Video title\n\nBy Creator\n\n[0:00] ..."
  },
  "metadata": {
    "source": "youtube-transcript",
    "videoId": "aircAruvnKk",
    "language": "en",
    "title": "Video title",
    "authorName": "Creator"
  },
  "credits": { "used": 5 }
}

Title, creator, duration, language, and thumbnail fields are included when the caption source provides them. You can also request HTML, raw HTML, links, a summary, or structured JSON through the usual scrape formats.

Price

Every successful YouTube transcript response costs 5 credits. Missing captions and failed requests are not charged.

Dedicated endpoint versus page scraping

POST /youtube/transcript always returns a transcript and does not accept browser actions, proxy controls, or waitFor. To load the YouTube page itself, use POST /scrape with browser options or use POST /screenshot to capture it.

For compatibility, POST /scrape still auto-detects supported YouTube video URLs when no browser action or waitFor value is present. The Playground has separate YouTube and Scrape screens so you can try both request paths.

Errors

Do not blindly retry 404: a video without available captions will keep returning the same result.