Add HF Spaces config

README.md

@@ -1,295 +1,9 @@
-# 🎬 Somira Content Automation
-
-**AI-powered video generation pipeline that transforms text scripts into professional product advertisements.**
-
 ---
-
-## 🚀 Quick Start
-
-### 1. Installation
-```bash
-# Clone and setup
-git clone <your-repo>
-cd somira-automation
-
-# Create virtual environment (recommended)
-python -m venv venv
-source venv/bin/activate  # Windows: venv\Scripts\activate
-
-# Install dependencies
-pip install -r requirements.txt
-```
-
-### 2. API Setup
-
-**You need these API keys:**
-
-#### Gemini API (Free)
-1. Go to https://aistudio.google.com/app/apikey
-2. Click "Create API Key"
-3. Copy the key
-
-#### RunwayML API ($10 minimum)
-1. Go to https://dev.runwayml.com/
-2. Sign up and create organization
-3. Go to "API Keys" → "Create new key"
-4. Add $10+ credits in "Billing" tab
-
-#### Google Cloud (Free tier available)
-1. Go to https://console.cloud.google.com/
-2. Create project → Enable "Text-to-Speech API"
-3. Create service account → Download JSON key
-4. Create storage bucket
-
-### 3. Configuration
-```bash
-# Copy and edit environment file
-cp .env.example .env
-```
-
-Edit `.env` with your keys:
-```bash
-# Required API Keys
-GEMINI_API_KEY=AIzaSyC_your_key_here
-RUNWAYML_API_KEY=key_your_runwayml_key_here
-GOOGLE_APPLICATION_CREDENTIALS=/path/to/service-account.json
-GCS_BUCKET_NAME=your-bucket-name
-
-```
-
-### 4. Verify Setup
-```bash
-python main.py --health-check
-```
-You should see: `✅ All systems operational!`
-
-### 5. Generate Your First Video
-```bash
-python main.py
-```
-
+title: Elvoro Tools
+emoji: 🚀
+colorFrom: blue
+colorTo: purple
+sdk: docker
+pinned: false
+app_port: 7860
 ---
-
-## 🎯 What It Does
-
-This system automatically creates 15-second vertical videos (perfect for TikTok/Reels) by:
-
-1. **AI Video Generation** - Creates unique hook videos using RunwayML Gen-4
-2. **Smart Content Selection** - Gemini AI analyzes your script to pick relevant product footage
-3. **Professional Voice-overs** - Converts text to natural speech using Google TTS
-4. **Auto Editing** - Merges videos, adds background music, subtitles, and effects
-5. **Cloud Storage** - Uploads final videos to Google Cloud Storage
-
-**Pipeline Time**: ~1-2 minutes per video
-
----
-
-## 💻 Usage
-
-### Basic Commands
-```bash
-# Generate video with default content
-python main.py
-
-# Test system (uses minimal credits)
-python main.py --test
-
-# Health check
-python main.py --health-check
-
-# Custom content
-python main.py --strategy strategy.json --script script.txt
-```
-
-### Custom Content
-Create `my_script.txt`:
-```
-I heard a pop and my neck was stuck. After one minute with Somira massager, the pain was gone. This product actually works!
-```
-
-Create `my_strategy.json`:
-```json
-{
-  "brand": "Somira",
-  "gemini_prompt": "A dramatic scene showing neck pain relief",
-  "runway_prompt": "Person experiencing neck pain then relief",
-  "style": "commercial",
-  "aspect_ratio": "9:16",
-  "duration": 5
-}
-```
-
-Run:
-```bash
-python main.py --strategy my_strategy.json --script my_script.txt
-```
-
-### Batch Processing
-```python
-import asyncio
-from automation import ContentAutomation
-
-async def create_videos():
-    automation = ContentAutomation(config)
-    
-    scripts = ["script1.txt", "script2.txt", "script3.txt"]
-    for script_file in scripts:
-        with open(script_file) as f:
-            result = await automation.execute_pipeline(strategy, f.read())
-        print(f"Created: {result['final_url']}")
-
-asyncio.run(create_videos())
-```
-
----
-
-## 💰 Pricing
-
-### Cost Per Video
-| Service | Cost |
-|---------|------|
-| RunwayML (5s video) | ~$0.50 |
-| Gemini API | ~$0.001 |
-| Google TTS | ~$0.001 |
-| Cloud Storage | ~$0.001 |
-| **Total** | **~$0.50** |
-
-### Monthly Estimate (100 videos)
-- **RunwayML**: $50
-- **Other services**: $2-5
-- **Total**: ~$55/month
-
----
-
-## 🏗️ How It Works
-
-### Pipeline Steps
-1. **Asset Generation** (30-60s)
-   - AI creates hook video from prompt
-   - Selects 3 relevant product videos
-   - Generates voice-over from script
-   - Picks background music
-
-2. **Video Composition** (10-20s)
-   - Merges all video clips
-   - Adds audio tracks and music
-   - Applies transitions
-
-3. **Subtitles** (5-10s)
-   - Generates animated subtitles
-   - Times them to voice-over
-
-4. **Cloud Upload** (5-15s)
-   - Uploads to Google Cloud Storage
-   - Returns public URL
-
-### Output Specifications
-- **Format**: MP4, H.264
-- **Aspect Ratio**: 9:16 (vertical)
-- **Duration**: 15 seconds max
-- **Resolution**: 1080x1920
-- **Audio**: 44.1kHz, stereo
-
----
-
-## 🔧 Technical Details
-
-### Project Structure
-```
-somira-automation/
-├── main.py                 # CLI entry point
-├── automation.py           # Pipeline orchestrator
-├── api_clients.py          # Gemini, RunwayML, TTS, GCS
-├── video_renderer.py       # Video processing engine
-├── asset_selector.py       # AI video selection
-├── utils.py                # Logging & utilities
-├── requirements.txt        # Python dependencies
-└── config/
-    ├── api_keys.yaml       # API configurations
-    └── content_strategies.yaml
-```
-
-### Key Dependencies
-- `moviepy` - Video editing and composition
-- `google-generativeai` - Gemini API client
-- `google-cloud-texttospeech` - TTS service
-- `google-cloud-storage` - Cloud storage
-- `aiohttp` - Async HTTP requests
-- `pandas` - Data processing
-
-### API Requirements
-- **Gemini**: Free tier available
-- **RunwayML**: $10 minimum deposit
-- **Google Cloud**: $300 free credits for new accounts
-- **Storage**: 5GB free tier
-
----
-
-## 🐛 Troubleshooting
-
-### Common Issues
-
-**"API key not found"**
-- Check `.env` file exists and has correct keys
-- Restart terminal after adding keys to `.env`
-
-**"Insufficient RunwayML credits"**
-- Add credits at https://dev.runwayml.com/
-- Minimum $10 required
-
-**"Google Cloud permission denied"**
-- Verify service account JSON path in `.env`
-- Check service account has "Storage Admin" role
-
-**"Module not found"**
-```bash
-pip install -r requirements.txt
-```
-
-**Videos taking too long**
-- RunwayML generation takes 30-60 seconds
-- Use `--test` for quick verification
-
-### Performance Tips
-- Keep scripts under 200 characters for optimal TTS
-- Use specific, visual prompts for better AI videos
-- Test with `--test` flag before full runs
-- Monitor API usage in respective dashboards
-
----
-
-## 📞 Support
-
-### Debugging
-- Run with `--verbose` for detailed logs
-- Check console output for specific error messages
-- Verify all APIs are enabled in their consoles
-
-### Cost Control
-- Use `--test` frequently during development
-- Set billing alerts in Google Cloud & RunwayML
-- Monitor usage in API dashboards
-
-### Security
-- ✅ Never commit `.env` file (included in `.gitignore`)
-- ✅ Use environment variables for all keys
-- ✅ Rotate API keys every 90 days
-- ❌ Never hardcode keys in source files
-
----
-
-## 🎉 Next Steps
-
-1. ✅ Complete API setup
-2. ✅ Run `python main.py --health-check`
-3. ✅ Test with `python main.py --test` 
-4. ✅ Generate first video with `python main.py`
-5. 🚀 Customize scripts and strategies for your products
-6. 📈 Scale with batch processing for multiple videos
-
-**Need help?** Check the error messages in console - they're designed to be helpful and specific about what went wrong.
-
----
-
-*Happy video generating! 🎬*
-..