Ever clicked on a 30-minute YouTube video only to realise it was a waste of time?
"Worth the Watch?" is a prototype AI-powered video summarisation tool that helps you decide whether to watch a video before you sink your time into it based on reading a (hopefully) condensed blurb
It uses Google Gemini AI and yt-dlp to:
- Extract video metadata (title, uploader, views, likes, etc.)
- Summarise the video based on its content
- Give you a quick preview so you can watch smarter!
- Enter a YouTube URL into the app.
- The system fetches metadata (title, uploader, views, etc.) using yt-dlp.
- If available, it extracts the transcript for more accurate summaries.
- It sends this data to Gemini AI, which generates a structured summary.
- You get a quick, AI-powered preview—so you can decide whether it's worth watching!
Component | Description |
---|---|
🧠 Google Gemini | Used for text summarisation and structured outputs. |
🎮 yt-dlp | Extracts YouTube metadata & transcripts. |
🌍 Streamlit | Simple & interactive frontend UI. |
🔑 dotenv (.env) | Stores API keys & credentials securely. |
🛂 requirements.txt | Lists all dependencies to run the project. |
pip install -r requirements.txt
- Go to Google AI Studio.
- Generate an API key and store it in a
.env
file like this:
GEMINI_API_KEY=your-api-key-here
streamlit run app.py
- Quick Video Previews – Summarises content before you watch.
- AI-Powered Insights – Extracts key topics, timestamps, and speakers.
- Metadata Extraction – Fetches views, likes, and uploader details.
- Saves You Time – or something like that