A multimodal data summarization application using Google's Gemini AI to summarize content from YouTube videos, images, and documents.
video.mp4
IMAGE.mp4
compressed.DOCUMENT-DEMO.mp4
- YouTube Video Summarizer: Extract transcripts and generate comprehensive summaries
- Image Analyzer & Summarizer: Analyze and describe images with AI
- Document Summarizer: Process PDF documents and generate summaries
When deploying to cloud platforms like Hugging Face, YouTube often detects automated tools like yt-dlp as bots and requires authentication, causing the error:
ERROR: [youtube] video_id: Sign in to confirm you're not a bot. Use --cookies-from-browser or --cookies for the authentication.
pip install -r requirements.txt- Run the Streamlit app:
streamlit run app.py- Navigate to the YouTube Video Summarizer section
- Paste a YouTube URL
- Click "Summarise Now"
Create a .env file with:
GOOGLE_API_KEY=your_google_api_key_here
streamlit==1.48.0google-generativeai==0.8.5youtube-transcript-api==0.6.2yt-dlp==2025.8.11python-dotenv==1.1.1- And more (see
requirements.txt)