Overview
Watching videos to extract specific data is slow and tiring. This workflow acts as a Video Distiller, automating the research process by scraping full transcripts and fetching official metadata. An AI expert then merges statistics and every spoken word into a structured report that highlights core arguments, target audience insights, and key takeaways.
Who This Is For
This template is ideal for those who need to digest deep, educational, or professional video content quickly.
- Market Researchers. Quickly analyze competitor content strategies and selling points.
- Students & Researchers. Extract core knowledge points from lectures or documentaries without manual note-taking.
- Content Creators. Get a detailed framework of existing videos to inspire scripts or blog posts.
- Corporate Teams. Convert training videos into written materials for easy archiving and retrieval.
How It Works
- ID Isolation
- The workflow extracts the unique video ID from your provided URL for API processing.
- Data Gathering
- It calls the official YouTube Data API to retrieve engagement stats, titles, and descriptions.
- Transcript Scraping
- Using an Apify scraper, the system pulls the full subtitle text directly from the video.
- AI Synthesis
- An AI model interprets the combined raw transcript and official data to generate a professional analysis report.
What You'll Need
- Google API Key. Obtain a Google API key from Google Cloud Console.
- Apify API Token. Find your personal token in the Apify Console.
- YouTube URL. The full link to the video you want to analyze.
How to Use
Step 1. Tap the "Test Run" button and enter your YouTube URL, Google API key, and Apify API.
Step 2. Click "Start Running".
Step 3. Once the workflow finishes, tap the "Generate Analysis Report" to view the report.
FAQs
Why are two different APIs needed?
The Google API provides metadata like views, while Apify scrapes the full transcript, which the official API does not easily provide.
What should I do if the workflow shows a "timeout exceeded" error?
Try running a video under 1 minute. If it succeeds, the problem is the platform's 60-second limit for Code nodes. You can fix this by changing the Python script in the "Scrape Video Transcript" node from a traditional polling method to a synchronous request using run-sync-get-dataset-items with the &wait=55 parameter.
Does this support non-English videos?
The scraper is set to English subtitles by default. For other languages, the Python code in the "Scrape Video Transcript" node can be modified.
What is the video length limit?
To save system resources, the scraping node has a 60-second runtime limit. While optimized, videos over 20-30 minutes may still trigger a timeout if the transcript is too large to process quickly.
What if a video has no subtitles?
If no transcript is found, the AI will build the report using the available official metadata and description.
