YouTube Transcript & Subtitles Scraper

What does it do?

The YouTube Transcript Scraper extracts transcripts, subtitles, and captions from any YouTube video in bulk -- no API key or YouTube Data API quota required. It supports auto-generated captions, manually uploaded subtitles, and any language, making it the most reliable tool for extracting spoken content from YouTube videos, playlists, channels, and Shorts at scale. Feed the output directly into LLMs, RAG pipelines, or content repurposing workflows.

What data does it extract?

• Full transcript text -- complete spoken content of the video as plain text
• Timed segments -- individual caption segments with start time, end time, and duration
• Video title -- the title of the YouTube video
• Channel name -- the channel that published the video
• Video URL -- direct link to the source video
• View count -- total number of views
• Upload date -- when the video was published
• Video duration -- total length of the video
• Language -- detected or selected transcript language
• Thumbnail URL -- video thumbnail image
• Description -- video description text

Use cases

1. RAG and LLM fine-tuning -- Extract transcripts from hundreds of educational or domain-specific YouTube videos to build a knowledge base for retrieval-augmented generation (RAG). Use the structured text to fine-tune language models on specialized topics like finance, medicine, or engineering.

2. Content repurposing at scale -- Convert YouTube video content into blog posts, social media threads, newsletters, or podcast show notes. Marketing teams use this to transform a single video into 10+ pieces of written content across platforms.

3. Video SEO and competitor analysis -- Analyze the spoken content of top-ranking YouTube videos in your niche. Identify keyword patterns, topic coverage, and content gaps to optimize your own video scripts and descriptions for better search rankings.

How to use

1. Navigate to the YouTube Transcript Scraper on Apify Store and click "Try for free."
2. In the URLs field, paste YouTube video URLs, playlist URLs, channel URLs, or raw video IDs. You can mix and match formats.
3. Select your preferred Language (default: English). The scraper will fall back to available languages if your preference is not available.
4. Choose an Output Format: full-text (plain text block), segments (timestamped chunks), or both.
5. Toggle Include Timestamps and Include Metadata as needed.
6. Click Start. Transcripts are extracted and saved to the Dataset tab.
7. Export results in JSON, CSV, or Excel, or integrate via API for automated pipelines.

Input parameters

Parameter	Type	Required	Description
urls	Array	Yes	YouTube video URLs, playlist URLs, channel URLs, or video IDs
language	String	No	Preferred transcript language code (default: en)
outputFormat	Enum	No	full-text, segments, or both (default: both)
includeTimestamps	Boolean	No	Include start/end times for each segment (default: true)
maxVideos	Integer	No	Maximum videos to process, up to 5,000 (default: 50)
includeMetadata	Boolean	No	Include video title, channel, views, etc. (default: true)
maxConcurrency	Integer	No	Concurrent requests, 1-20 (default: 5)
proxyConfiguration	Object	No	Apify Proxy country routing only. The actor always uses BUYPROXIES94952; custom proxy URLs and alternate groups are ignored

Output example

{
  "videoUrl": "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
  "title": "How to Build a RAG Pipeline in 2026",
  "channel": "AI Engineering Academy",
  "viewCount": 245000,
  "uploadDate": "2026-02-15",
  "duration": "14:32",
  "language": "en",
  "fullText": "Welcome to this tutorial on building a retrieval-augmented generation pipeline. Today we'll cover vector databases, embedding models, and...",
  "segments": [
    {
      "text": "Welcome to this tutorial on building a retrieval-augmented generation pipeline.",
      "start": 0.0,
      "end": 4.2,
      "duration": 4.2
    },
    {
      "text": "Today we'll cover vector databases, embedding models, and chunking strategies.",
      "start": 4.2,
      "end": 8.8,
      "duration": 4.6
    }
  ],
  "thumbnailUrl": "https://i.ytimg.com/vi/dQw4w9WgXcQ/maxresdefault.jpg"
}

Pricing

• Start event: $0.005 per run
• Per transcript: $0.004 per video transcript extracted

Approximate cost: $4 per 1,000 transcripts. No API key, no YouTube Data API quota, and no monthly subscription -- pay only for successful extractions.

FAQ

Q: Do I need a YouTube Data API key? A: No. This scraper works without any API key or Google account. It extracts transcripts directly, bypassing YouTube API quotas entirely.

Q: Does it work with auto-generated captions? A: Yes. The scraper handles both manually uploaded subtitles and YouTube's auto-generated captions in any language.

Q: Can I scrape entire playlists or channels? A: Yes. Pass a playlist URL or channel URL and the scraper will automatically discover and process all videos, up to your configured maxVideos limit.

Q: What languages are supported? A: All languages that YouTube provides transcripts for are supported. Set your preferred language code and the scraper will use it if available, or fall back to the best available alternative.

Q: How do I handle geo-restricted videos? A: Use the proxyConfiguration.countryCode parameter to route requests through Apify Proxy in the appropriate country. The actor always enforces the BUYPROXIES94952 proxy group for reliability.

Q: Can I use this for LLM training data? A: Yes. The full-text output format is ideal for LLM fine-tuning datasets. Process up to 5,000 videos per run to build large-scale training corpora.

Why choose this over alternatives?

• No API key needed -- Zero setup friction. No Google Cloud project, no API quota limits, no OAuth tokens.
• Massive scale -- Process up to 5,000 videos per run with configurable concurrency up to 20 parallel requests.
• Multiple input formats -- Videos, playlists, channels, Shorts, and raw video IDs all accepted in a single run.
• LLM-ready output -- Full-text and segmented formats designed for direct ingestion into RAG pipelines and fine-tuning workflows.
• 97%+ success rate -- Proven across 327+ runs with 52 users and 28 active users in the last 7 days.
• Auto-generated caption support -- Works even when video creators haven't uploaded manual subtitles.