Youtube Video Summarizer

An Apify actor that extracts YouTube video metadata, transcripts, and generates AI-powered summaries using OpenAI's GPT models.

Why Choose Us?

• AI-Powered Summaries: Generate intelligent, contextual summaries of YouTube videos
• Comprehensive Data Extraction: Get video metadata, transcripts, channel information, and engagement metrics
• Robust Proxy Handling: Automatic fallback to residential proxies when needed
• Bulk Processing: Process multiple videos in a single run
• Multiple Output Formats: Get both short and detailed markdown summaries

Key Features

• 🎯 Smart Transcript Extraction: Automatically finds and extracts video transcripts
• 🤖 AI Summarization: Uses OpenAI GPT-4o-mini for intelligent content summarization
• 📊 Rich Metadata: Extracts views, likes, comments, channel info, and more
• 🌐 Proxy Support: Built-in proxy handling with automatic fallback
• ⚡ Fast Processing: Efficient processing with connection pooling
• 🔄 Error Recovery: Robust error handling and retry mechanisms
• 📱 Mobile-Friendly: Works with all YouTube video formats

Input

Required Fields

• startUrls (array): List of YouTube video URLs to process

  {
    "startUrls": [
      { "url": "https://www.youtube.com/watch?v=Z4hVGCWH1Kc" },
      { "url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ" }
    ]
  }
  

Optional Fields

• proxyConfiguration (object): Proxy settings for requests

  {
    "proxyConfiguration": {
      "useApifyProxy": false
    }
  }
  

• maxVideoDuration (integer): Maximum video duration in seconds (default: 3600)

• language (string): Preferred language for transcript extraction (default: "en")

Output

The actor outputs structured data to the dataset with the following fields:

{
  "url": "https://www.youtube.com/watch?v=VIDEO_ID",
  "id": "VIDEO_ID",
  "title": "Video Title",
  "description": "Video description",
  "published_date": "2024-01-15",
  "relativeDate": "2 days ago",
  "duration_seconds": 1200,
  "views": 50000,
  "likes": 2500,
  "comment_count": 150,
  "summary": "AI-generated short summary (3-4 sentences)",
  "long_summary": "# Detailed Markdown Summary\n\nStructured summary with headings and sections",
  "channel": {
    "id": "CHANNEL_ID",
    "name": "Channel Name",
    "handle": "@channelhandle",
    "url": "https://www.youtube.com/@channelhandle",
    "subscriberCount": "1M subscribers",
    "logo": [
      {
        "url": "https://...",
        "width": 176,
        "height": 176
      }
    ],
    "badges": ["Verified"]
  },
  "success": true,
  "error": null
}

🚀 How to Use the Actor (via Apify Console)

1. Log in at https://console.apify.com and go to Actors
2. Find the youtube-video-summarizer actor and click it
3. Configure inputs:
   • Add YouTube video URLs in the startUrls field
   • Set proxy configuration if needed
   • Adjust max video duration and language preferences
4. Run the actor by clicking the Start button
5. Monitor logs in real-time to see processing progress
6. Access results in the OUTPUT tab
7. Export results to JSON, CSV, or other formats

Best Use Cases

• Content Research: Quickly summarize multiple videos for research purposes
• Content Creation: Generate summaries for blog posts or articles
• Market Analysis: Analyze competitor content and messaging
• Educational Content: Create study materials from educational videos
• Social Media: Generate engaging social media posts from video content
• SEO Content: Create meta descriptions and summaries for video content

Proxy Configuration

The actor includes intelligent proxy handling:

• Default: No proxy (direct connection to YouTube)
• Automatic Fallback: If YouTube blocks the request, automatically switches to residential proxy
• Retry Logic: Up to 3 retries with residential proxy if needed
• Sticky Fallback: Once fallback occurs, uses residential proxy for remaining requests
• Clear Logging: All proxy events are logged for transparency

Technical Details

Dependencies

• apify-client: Apify platform integration
• openai: OpenAI API for AI summarization
• yt-dlp: YouTube video data extraction
• requests: HTTP requests with proxy support
• beautifulsoup4: HTML parsing for channel data
• python-dotenv: Environment variable management

Processing Flow

1. Input Validation: Validates URLs and configuration
2. Proxy Setup: Initializes proxy configuration
3. Video Extraction: Uses yt-dlp to extract video metadata
4. Transcript Extraction: Finds and downloads video transcripts
5. Channel Data: Extracts channel information and logos
6. AI Summarization: Generates summaries using OpenAI GPT
7. Data Formatting: Structures output for dataset storage
8. Error Handling: Graceful error handling with detailed logging

Error Handling

• Video Duration: Skips videos exceeding max duration
• Missing Transcripts: Handles videos without available transcripts
• API Errors: Retries OpenAI API calls with exponential backoff
• Proxy Failures: Automatic fallback to residential proxies
• Network Issues: Robust timeout and retry mechanisms

Frequently Asked Questions

Q: Do I need an OpenAI API key? A: No, the actor uses a built-in OpenAI API key for generating AI-powered summaries.

Q: What happens if a video has no transcript? A: The actor will skip the video and log an error, but continue processing other videos.

Q: Can I process private videos? A: No, only publicly available videos can be processed.

Q: How long does processing take? A: Depends on video length and transcript availability. Typically 30-60 seconds per video.

Q: What if YouTube blocks my requests? A: The actor automatically switches to residential proxies and retries up to 3 times.

Q: Can I customize the summary length? A: Currently, summaries are optimized for 3-4 sentences (short) and detailed markdown (long).

Support and Feedback

• Documentation: Apify Academy
• Community: Apify Discord
• Issues: Report bugs and feature requests through the Apify platform

Cautions

• Data Collection: Only collects data from publicly available YouTube videos
• Legal Compliance: Users are responsible for ensuring compliance with YouTube's Terms of Service
• API Usage: OpenAI API usage is included with the actor
• Rate Limits: Respect YouTube's rate limits and terms of service
• Privacy: No private or password-protected content is accessed

---

Note: This actor is designed for educational and research purposes. Always respect YouTube's Terms of Service and applicable laws when scraping content.