YouTube Transcript & Metadata Scraper (RAG) avatar

YouTube Transcript & Metadata Scraper (RAG)

Pricing

$1.50 / 1,000 transcript extracteds

Go to Apify Store
YouTube Transcript & Metadata Scraper (RAG)

YouTube Transcript & Metadata Scraper (RAG)

Ultra-fast YouTube Transcript & Metadata Extractor via internal Innertube API. Extract timestamped transcripts, plain text, LLM-ready Markdown, chapters, views, and channel data from URLs, Video IDs, or Search Queries. 80% cheaper ($1.25/1k).

Pricing

$1.50 / 1,000 transcript extracteds

Rating

0.0

(0)

Developer

Cedric poma

Cedric poma

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

a day ago

Last modified

Share

YouTube Transcript, Chapters & Metadata Scraper (AI / RAG Ready)

Extract YouTube video transcripts, timestamps, chapters, and full video metadata in milliseconds using YouTube's lightweight internal Innertube API — no headless browser required.

Built specifically for AI Agents, RAG pipelines, n8n, Make, Zapier, and bulk research workflows.


Why Use This YouTube Transcript Scraper?

FeatureLegacy Browser Scrapers ($5 - $10 / 1k)YouTube Transcript Pro ($1.25 / 1k)
Price per 1,000 Videos$5.00 – $10.00$1.25 / 1,000 results
Execution Speed8 – 25 seconds per video< 0.4 seconds per video
Memory Footprint2,048 MB – 4,096 MB (Chromium)128 MB RAM (Pure HTTP)
Input FlexibilityVideo URLs onlyVideo URLs, Shorts, Video IDs + Keyword Search
AI / LLM OutputRaw text onlyClean Text + Timestamped JSON + RAG Markdown + Chapters

Key Features

  • Direct URLs, Shorts & Video IDs: Pass any standard youtube.com/watch?v=..., youtu.be/..., youtube.com/shorts/..., or raw 11-character Video ID.
  • Built-in Keyword Search: Pass searchQueries (e.g., ["AI agents tutorial", "n8n automation"]) to automatically discover top videos and extract their full transcripts in a single run.
  • Multi-Language Support: Specify your preferredLanguage (en, fr, es, de, pt, ja, etc.) with automatic fallback to manual or auto-generated (asr) captions.
  • Description Chapter Parsing: Automatically extracts chapters (00:00 Intro, 02:15 Architecture, etc.) from the video description.
  • LLM / RAG Markdown Output: Generates a structured ragMarkdown field ready to feed directly into OpenAI, Claude, Gemini, Pinecone, Qdrant, or n8n AI nodes.

Input Example

{
"videoUrls": [
"https://www.youtube.com/watch?v=dQw4w9WgXcQ"
],
"searchQueries": [
"mcp server tutorial"
],
"maxSearchResultsPerQuery": 3,
"preferredLanguage": "en",
"includeTimestamps": true,
"includeMarkdown": true
}

Output Example

{
"videoId": "dQw4w9WgXcQ",
"url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
"title": "Rick Astley - Never Gonna Give You Up (Official Video) (4K Remaster)",
"channelName": "Rick Astley",
"channelId": "UCuAXFkgsw1L7xaCfnd5JJOw",
"publishDate": "2009-10-25T06:57:33-07:00",
"durationSeconds": 213,
"durationFormatted": "03:33",
"viewCount": 1600000000,
"hasTranscript": true,
"language": "en",
"isAutoGenerated": false,
"wordCount": 412,
"fullText": "♪ We're no strangers to love ♪ ♪ You know the rules and so do I ♪ ...",
"chapters": [],
"segments": [
{
"start": 18.64,
"duration": 3.24,
"timestamp": "00:18",
"text": "♪ We're no strangers to love ♪"
}
],
"ragMarkdown": "# Rick Astley - Never Gonna Give You Up...\n## Transcript\n[00:18] ♪ We're no strangers to love ♪..."
}