🎥 AI Video Summarizer

An AI-powered Streamlit app that transcribes and summarizes videos using local machine learning models. Supports YouTube links (only locally) and uploaded video files, with options to export results as PDF or plain text.

✨ Features

• 📺 YouTube & Local Uploads – Summarize videos from YouTube URLs or uploaded files
• 🎙 Speech-to-Text with Whisper – Local transcription in multiple languages
• 📝 Flexible Summaries – Choose between key points, paragraphs, or detailed outputs
• ⏱ Timestamps & Chunks – Breaks transcripts into manageable chunks with summaries per section
• 📄 Export Options – Download summaries as PDF reports or raw transcripts as text
• 🚀 Runs Locally – No reliance on paid APIs; everything uses open-source models

🛠 Tech Stack

• Streamlit – Web UI
• OpenAI Whisper – Speech-to-text transcription
• Hugging Face Transformers – Local summarization with pretrained models (BART, T5)
• yt-dlp – YouTube video/audio downloader
• MoviePy – Video/audio extraction
• ReportLab – PDF export

📦 Installation

# Clone repo
git clone https://github.com/your-username/ai-video-summarizer.git
cd ai-video-summarizer

# Create virtual environment (recommended)
python -m venv venv
source venv/bin/activate   # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Dependencies

streamlit
torch
transformers
openai-whisper
moviepy
yt-dlp
reportlab

🚀 Usage

1. Start the app:

   streamlit run video_summarizer.py

2. Load models via the sidebar (choose Whisper size + summarizer)

3. Select input method: YouTube URL or Upload Video

4. Click Start Processing to transcribe & summarize

5. View results in the Results tab:

   • Executive summary
   • Timestamped breakdowns
   • Full transcript

6. Export to PDF or TXT

🌐 Deployment Options

Online Demo: [https://kmaanasar-ai-video-summarizer-video-summarizer-lgwh3m.streamlit.app/]

• ✅ File upload works perfectly
• ⚠️ YouTube downloads may be limited due to cloud restrictions

Local Installation: (Recommended for full YouTube support)

• Follow installation steps above
• All features work reliably

📂 Project Structure

├── app.py                 # Main Streamlit app
├── requirements.txt       # Dependencies
├── packages.txt           # Streamlit Dependencies
├── README.md              # Documentation

⚖️ License

MIT License. Free to use and modify.