Video Transcript Scraper: Youtube, TikTok, Facebook, Loom, etc

Universal Video Transcript Scraper 🎬

Universal Video Transcript Scraper, your all-in-one video transcript scraper and video transcript downloader for YouTube, TikTok, Facebook, and Loom. Extract clean, timestamped transcripts from public videos across multiple platforms—perfect for content creators, researchers, educators, and developers.

---

🌟 Why Choose Universal Video Transcript Scraper?

Feature	Benefit
✅ Multi-Platform Support	Works with YouTube, TikTok, Facebook, and Loom URLs seamlessly
✅ Automatic Platform Detection	No need to specify source—scraper auto-detects and adapts
✅ Unified or Platform-Specific Output	Choose “unified” format for consistency or “platform-specific” for detailed fields
🛡️ Smart Proxy Fallback	Automatically falls back from direct → datacenter → residential proxies if blocked
🔄 Intelligent Retry Mechanism	Up to 3 retries per video per proxy type with exponential backoff
💾 Real-Time Data Saving	Immediate result persistence ensures no data loss on interruption

---

🔑 Key Features

• ✅ Automatic Platform Detection Detects whether a URL belongs to YouTube, TikTok, Facebook, or Loom, then applies the correct scraping logic.
• 🌐 Unified Output Format Provides a consistent JSON schema (url, platform, video_id, transcript_text, transcript_segments, success, timestamp) for easy downstream processing.
• 📁 Platform-Specific Format (Optional) Output includes fields unique to each platform (e.g., duration for YouTube segments or start_time for Loom) when “platform-specific” is selected.
• 🛡️ Proxy Resilience If the platform blocks requests, automatically rotates through direct, datacenter, and residential proxies to maximize success.
• 🔄 Advanced Retry Logic Retries each video up to 3 times per proxy type with exponential backoff delays to avoid rate limits.
• ⚡ Fast & Reliable Uses efficient scraping strategies (e.g., official transcript APIs, headless browsers) and intelligent delays to scrape transcripts in 1–3 seconds per video.
• 📊 Structured Transcript Segments Each transcript segment includes text, start (or start_time), and (where available) duration.
• 💾 Live Data Persistence Saves each result to the default dataset as soon as it’s ready, ensuring no data loss even if the run is interrupted.

---

📝 Input

Save an input.json file with the following structure:

{
  "outputFormat": "unified",
  "proxyConfiguration": {
    "useApifyProxy": false
  },
  "startUrls": [
    {
      "url": "https://www.youtube.com/watch?v=e-ORhEE9VVg",
      "method": "GET"
    },
    {
      "url": "https://www.loom.com/share/e41353f2fe1c43eba6c6829693e0f2c5",
      "method": "GET"
    },
    {
      "url": "https://www.facebook.com/briantylercohen/videos/1201813848348387",
      "method": "GET"
    },
    {
      "url": "https://www.tiktok.com/@david.webdeveloper/video/7506519006854253847",
      "method": "GET"
    }
  ]
}

Input Fields

Key	Type	Description
startUrls	array	List of video URLs from YouTube, TikTok, Facebook, or Loom. Each entry should include "url" and optional "method" (default: GET).
proxyConfiguration	object	Proxy settings—if useApifyProxy is false, the scraper first tries direct connection, then datacenter, then residential proxies on failure.
useApifyProxy	boolean	true to enable Apify proxy (recommended for high success against strict rate limits); false for automatic fallback logic.
outputFormat	string	Choose between "unified" (default) or "platform-specific" to tailor JSON fields per platform.

---

📦 Output

Upon successful execution, the actor will produce one JSON object per video in the default dataset. Below is a sample unified output:

{
  "url": "https://www.youtube.com/watch?v=e-ORhEE9VVg",
  "platform": "youtube",
  "video_id": "e-ORhEE9VVg",
  "transcript_text": "♪♪♪ ♪ NICE TO MEET YOU ♪ ...",
  "transcript_segments": [
    {
      "text": "♪♪♪",
      "start": 3.4,
      "duration": 4.866
    },
    {
      "text": "♪ NICE TO MEET YOU ♪",
      "start": 8.3,
      "duration": 1.266
    }
    // … more segments
  ],
  "success": true,
  "timestamp": "2025-06-02T07:56:00"
}

Note: For Loom, transcript_segments entries use start_time instead of start, but both fields remain consistent in the unified schema.

Output Fields (Unified)

Field	Type	Description
url	string	Original video URL as provided in startUrls.
platform	string	Detected platform: youtube, tiktok, facebook, or loom.
video_id	string	Extracted video identifier (e.g., e-ORhEE9VVg for YouTube, e41353f2fe1c43eba6c6829693e0f2c5 for Loom).
transcript_text	string	Complete transcript as a single concatenated string (easy for search or content repurposing).
transcript_segments	array	Array of segment objects. Each has: • text (string) • start or start_time (number, seconds) • duration (number, seconds; when available)
success	boolean	true if transcription succeeded; false if scraping failed or no transcript available.
timestamp	string	ISO-format timestamp (UTC) when the transcript was scraped (e.g., "2025-06-02T07:56:00").

If outputFormat is set to platform-specific, additional fields (e.g., duration for YouTube segments) will be included where supported.

---

🚀 How to Use (via Apify Console)

1. Open Apify Console Log in at https://console.apify.com and navigate to the Actors section.

2. Find Your Actor Search for video-transcript-scraper and click its card to open the details page.

3. Configure Input

   • In the INPUT tab, paste your input.json.
   • For Proxy configuration, enable Use Apify Proxy if desired; otherwise it will automatically fall back from direct → datacenter → residential on failures.
   • Add one or more video URLs (YouTube, Loom, Facebook, TikTok).

4. Choose Output Format

   • Select either Unified Format (consistent fields across all platforms) or Platform-Specific Format (detailed per-platform fields).

5. Run the Actor Click Run in the top-right. The actor will begin scraping transcripts and show real-time logs (including platform detection, proxy usage, and retry attempts).

6. Monitor Progress Observe logs to see when each URL is successfully scraped or if any errors occur.

7. Access Results Once complete, switch to the OUTPUT tab and open the Universal Video Transcript Data dataset.

8. Export Data Filter, preview, and export scraped transcripts to JSON or CSV for further analysis or integration.

That’s it! No coding required—leverage this video transcript downloader to batch-process videos from multiple platforms in one run.

---

🎯 Advanced Features

1. Smart Proxy Management

• Automatic Fallback: Starts with direct connection; if blocked, switches to Datacenter (AUTO), then to Residential proxies.
• User-Controlled Proxy Preference: If useApifyProxy: true, you can specify preferred Apify Proxy Groups like ["RESIDENTIAL"] or ["AUTO"].

2. Intelligent Platform Detection

• Detects YouTube, TikTok, Facebook, or Loom based on URL patterns.
• Applies platform-specific scraping logic automatically (e.g., official YouTube API if available, headless browsing for Loom).

3. Robust Error Handling

• 3 Retries per Proxy Type: Each video gets 3 attempts per proxy (direct, datacenter, residential).
• Exponential Backoff Delays: Delays grow after each retry to respect rate limits.
• Graceful Degradation: A failure on one video does not stop processing of remaining URLs.

4. Unified vs. Platform-Specific Output

• Unified Format: Consistent JSON schema across all platforms—ideal for pipelines or search indexing.
• Platform-Specific Format: Retains all original fields (e.g., YouTube’s duration per segment, TikTok’s auto-generated captions, Loom’s start_time).

---

💡 Best Use Cases

• ✍️ Content Repurposing Convert video dialogues into blog posts, articles, or social media micro-content.
• 📊 SEO & Keyword Research Use extracted transcripts to identify trending keywords, optimize on-page SEO, or analyze competitor video strategies.
• 🧠 Academic & Market Research Gather data from educational talks, interviews, or product demos across YouTube, TikTok, Facebook, and Loom for sentiment analysis.
• 🔈 Accessibility Enhancement Generate text versions of video content for hearing-impaired audiences or subtitle generation.
• 🤖 Machine Learning & AI Training Collect large corpora of spoken language transcripts to train speech-to-text models, chatbots, or sentiment classifiers.
• 📚 E-Learning & Summarization Extract lecture transcripts or tutorial dialogues for generating study notes, summaries, or flashcards.

---

🔧 Technical Specifications

Supported Video Types & Platforms

Platform	URL Formats
YouTube	• Standard: https://www.youtube.com/watch?v=VIDEO_ID • Shortened: https://youtu.be/VIDEO_ID • Shorts: https://www.youtube.com/shorts/VIDEO_ID
Loom	• Public share link: https://www.loom.com/share/VIDEO_ID
Facebook	• Public watch link: https://www.facebook.com/watch/?v=VIDEO_ID
TikTok	• Standard: https://www.tiktok.com/@username/video/VIDEO_ID

Proxy Support

• ✅ Direct Connections: First attempt uses no proxy.
• ✅ Datacenter Proxies (AUTO): Automatic fallback upon direct-connection failure.
• ✅ Residential Proxies: Final fallback if datacenter proxies are blocked.
• ✅ User-Configurable: Enable Apify Proxy and specify groups (e.g., ["RESIDENTIAL"] or ["AUTO"]).

Performance

• Processing Speed: ~1–3 seconds per video (depending on transcript length and proxy delays).
• Batch Size: No upper limit—process unlimited URLs in one run.
• Success Rate: > 95% for public videos with available transcripts.
• Retries: Up to 6 attempts per video (3 per proxy type).

---

🤝 Support & Feature Requests

Need help or want to suggest improvements?

• 🐞 Bug Reports: Please submit an issue on the Apify community with the video URL and error details.
• 💡 Feature Requests: Want bulk export to Google Sheets, speaker diarization, subtitle formatting, or support for new platforms? Let us know!
• 📧 Contact: Reach out via scrapermindapi@gmail.com.

Your feedback drives ongoing improvements to this video transcript scraper and video transcript downloader. Thank you for using Universal Video Transcript Scraper!