Add Space metadata to README

README.md

@@ -1,268 +1,19 @@
-# 🤖 AI Article Summarizer with Text-to-Speech
-
-A beautiful web application that scrapes articles from URLs, generates concise summaries using Qwen3-0.6B, and optionally converts them to speech using Kokoro TTS.
-
-## ✨ Features
-
-- **🌐 Web Scraping**: Extract clean article text from any URL
-- **🤖 AI Summarization**: Generate concise summaries using Qwen3-0.6B
-- **🎵 Text-to-Speech**: Convert summaries to natural speech with Kokoro TTS
-- **🎭 Multiple Voices**: Choose from 8 different voice options
-- **📱 Responsive Design**: Works on desktop and mobile devices
-- **⚡ Real-time Processing**: Live status updates and progress indicators
-
-## 🚀 Quick Start
-
-### 1. Clone/Download Files
-
-Create a new directory and save these files:
-- `app.py` - Flask web application
-- `templates/index.html` - Web interface
-- `setup_web_app.sh` - Setup script
-
-### 2. Run Setup Script
-
-```bash
-# Make setup script executable
-chmod +x setup_web_app.sh
-
-# Run setup (installs everything automatically)
-./setup_web_app.sh
-```
-
-### 3. Start the Web Application
-
-```bash
-# Activate virtual environment
-source venv/bin/activate
-
-# Run the Flask app
-python app.py
-```
-
-### 4. Open in Browser
-
-Navigate to: **http://localhost:5000**
-
-## 📋 Manual Setup (Alternative)
-
-If you prefer manual setup:
-
-### Prerequisites
-
-- Python 3.8+
-- macOS: `brew install espeak`
-- Linux: `sudo apt-get install espeak-ng`
-
-### Installation
-
-```bash
-# Create virtual environment
-python3 -m venv venv
-source venv/bin/activate
-
-# Install packages
-pip install Flask torch transformers trafilatura soundfile kokoro librosa
-
-# Create directories
-mkdir -p templates static/audio static/summaries
-```
-
-## 🎭 Available Voices
-
-| Voice | Type | Quality | Description |
-|-------|------|---------|-------------|
-| af_heart ❤️ | Female | A | Warm, best quality (default) |
-| af_bella 🔥 | Female | A- | Energetic, high quality |
-| af_nicole 🎧 | Female | B- | Professional tone |
-| am_michael | Male | C+ | Clear male voice |
-| am_fenrir | Male | C+ | Strong male voice |
-| af_sarah | Female | C+ | Gentle female voice |
-| bf_emma 🇬🇧 | Female | B- | British accent |
-| bm_george 🇬🇧 | Male | C | British male accent |
-
-## 🖥️ Web Interface Features
-
-### Main Interface
-- Clean, modern design with gradient background
-- Real-time model loading status
-- URL input with validation
-- Optional text-to-speech toggle
-- Voice selection with quality indicators
-
-### Processing
-- Live progress indicators
-- Error handling with user-friendly messages
-- Summary statistics (compression ratio, word count)
-- Timestamp tracking
-
-### Results
-- Beautiful summary display with syntax highlighting
-- Integrated audio player for TTS output
-- Downloadable audio files
-- Responsive design for mobile devices
-
-## 📊 Technical Details
-
-### AI Models Used
-- **Qwen3-0.6B**: 600M parameter language model for summarization
-- **Kokoro TTS**: 82M parameter text-to-speech model
-- **Trafilatura**: Web scraping and content extraction
-
-### Performance
-- **Model Loading**: ~1-2 minutes on first startup
-- **Summarization**: ~2-5 seconds per article
-- **TTS Generation**: ~1-3 seconds for typical summaries
-- **Memory Usage**: ~2-4GB RAM (depending on hardware)
-
-### File Structure
-```
-your-project/
-├── app.py                 # Flask web application
-├── templates/
-│   └── index.html         # Web interface
-├── static/
-│   ├── audio/            # Generated audio files
-│   └── summaries/        # Cached summaries (optional)
-├── venv/                 # Virtual environment
-└── requirements.txt      # Python dependencies
-```
-
-## 🔧 Configuration
-
-### Environment Variables (Optional)
-```bash
-export FLASK_ENV=development    # For debugging
-export FLASK_PORT=5000         # Custom port
-```
-
-### Model Configuration
-- Models are loaded automatically on startup
-- First run downloads ~1.2GB of model files
-- Models are cached locally for faster subsequent starts
-
-## 🐛 Troubleshooting
-
-### Common Issues
-
-**Models not loading**
-```bash
-# Check internet connection and disk space
-df -h
-ping huggingface.co
-```
-
-**espeak not found**
-```bash
-# macOS
-brew install espeak
-
-# Linux
-sudo apt-get install espeak-ng
-```
-
-**Permission errors**
-```bash
-# Make sure virtual environment is activated
-source venv/bin/activate
-which python  # Should show venv path
-```
-
-**Port already in use**
-```bash
-# Kill process using port 5000
-lsof -ti:5000 | xargs kill -9
-
-# Or use different port
-python app.py --port 5001
-```
-
-### Performance Optimization
-
-**For faster startup:**
-- Keep models in memory between requests
-- Use SSD storage for model cache
-- Ensure sufficient RAM (4GB+ recommended)
-
-**For better quality:**
-- Use `af_heart` or `af_bella` voices
-- Keep summaries under 500 words for best TTS quality
-- Use clean, well-formatted article URLs
-
-## 📱 Usage Tips
-
-1. **Best Article Sources**: News sites, blogs, Wikipedia work well
-2. **URL Format**: Use full URLs (https://example.com/article)
-3. **Summary Length**: Typically 100-300 words from longer articles
-4. **Audio Quality**: Higher-grade voices sound more natural
-5. **Mobile Use**: Interface is fully responsive
-
-## 🔒 Privacy & Security
-
-- No article content is stored permanently
-- Audio files are generated locally
-- No data is sent to external services (except model downloads)
-- All processing happens on your machine
-
-## 🚀 Advanced Usage
-
-### API Endpoints
-
-The web app exposes these endpoints:
-
-- `GET /` - Main interface
-- `GET /status` - Model loading status
-- `GET /voices` - Available voice list
-- `POST /process` - Process article (JSON)
-
-### Example API Usage
-```javascript
-fetch('/process', {
-    method: 'POST',
-    headers: {'Content-Type': 'application/json'},
-    body: JSON.stringify({
-        url: 'https://example.com/article',
-        generate_audio: true,
-        voice: 'af_heart'
-    })
-});
-```
-
-## 📈 Future Enhancements
-
-Possible improvements:
-- Multiple language support
-- Custom voice training
-- Batch processing
-- Summary length control
-- Export options (PDF, EPUB)
-- Integration with read-later apps
-
-## 🤝 Contributing
-
-Feel free to:
-- Report bugs
-- Suggest features
-- Submit improvements
-- Share cool use cases
-
-## 📄 License
-
-This project uses:
-- Qwen3-0.6B (Apache 2.0)
-- Kokoro TTS (Apache 2.0)
-- Flask (BSD)
-- Other open-source libraries
-
-## 🙏 Acknowledgments
-
-- **Qwen Team** for the summarization model
-- **hexgrad** for Kokoro TTS
-- **Trafilatura** for web scraping
-- **Flask** for the web framework
-
 ---
+title: AI Article Summarizer (Qwen + Kokoro)
+emoji: 📰
+colorFrom: indigo
+colorTo: purple
+sdk: docker
+pinned: false
+license: apache-2.0
+short_description: Dark, fluid Flask app that scrapes articles, summarizes with Qwen 0.6B, and optionally speaks results with Kokoro TTS. CPU-only, no accounts needed.
+---
+
+# AI Article Summarizer
 
-**Enjoy your AI-powered article summarizer!** 🎉
+Dark, fluid Flask app that scrapes articles, summarizes with **Qwen 0.6B**, and optionally speaks results with **Kokoro TTS**.  
+Runs on the free CPU tier. First run downloads models; later requests are fast.
 
-For support or questions, check the troubleshooting section or open an issue.
+## Tips
+- If a site blocks scraping, paste the text or set `ALLOW_PROXY_FALLBACK=1` in Space settings.
+- Status endpoint: `/status`.