Offline-Translator (Powered by Seed-X)

A privacy-focused offline AI translation application built with Transformers and Tkinter, providing secure local translation without sending data to external servers.

⚠️ IMPORTANT NOTICE: This repository contains placeholder model files. You must download the actual model files (~14GB) before running the application. See Installation section for details.

✨ Key Features

• 🤖 Local AI Translation - Complete privacy protection with offline processing
• 🌍 Multi-language Support - Bidirectional translation between 8 languages (Chinese, English, Spanish, French, German, Japanese, Korean, Russian)
• ⚡ GPU Acceleration - Automatic GPU detection with CUDA support for faster translation
• 🎨 Modern Interface - Clean, intuitive graphical user interface
• ⌨️ Keyboard Shortcuts - Rich hotkey support for efficient workflow
• 📋 One-click Copy - Easy result copying to clipboard
• 🔄 Quick Language Swap - Instant source/target language switching
• 📏 Smart Text Limits - 5000 character limit with helpful user guidance

🔧 System Requirements

• Python: 3.8 or higher
• PyTorch: Latest stable version
• Transformers: Hugging Face transformers library
• Accelerate: For optimized model loading
• Tkinter: GUI framework (usually included with Python)

📦 Installation

1. Clone the repository:

git clone https://github.com/your-username/offline-translator.git
cd offline-translator

2. Install dependencies:

pip install -r requirements.txt

3. ⚠️ IMPORTANT: Download Model Files

   The repository contains placeholder model files. You MUST download the actual model before running the application.

   • 📁 Navigate to the models/ directory
   • 📖 Read the README.txt file for detailed download instructions
   • 🔗 Download the model files (~14GB) from the provided links
   • 📋 Replace the placeholder model.safetensors (0KB) with the actual model file

   Quick Setup:

   # After cloning, check the models directory
   cd models/
   cat README.txt  # Read download instructions
   
   # Download model files from the provided links
   # Replace model.safetensors with the downloaded file (~14GB)

4. Verify Installation:

   • Ensure model.safetensors is approximately 14GB (not 0KB)
   • All JSON configuration files should contain actual data
   • Run the application to test: python translator.py

🚀 Usage

1. Launch the application:

python translator.py

2. Setup process:
   • Select your preferred device (GPU recommended for speed)
   • Click "Load Model" to initialize the translation engine
   • Choose source and target languages
   • Enter text (up to 5000 characters)
   • Click "Translate" or press Ctrl+Enter

⌨️ Keyboard Shortcuts

Shortcut	Action
Ctrl+Enter	Translate text
Ctrl+L	Load model
Ctrl+Shift+C	Copy translation result
Ctrl+Shift+X	Clear input text
Ctrl+Shift+S	Swap source and target languages
F1	Show help information

🖥️ Interface Features

Device Selection

• Automatic GPU Detection - Scans for available CUDA devices
• GPU Priority - Graphics cards listed first, CPU as fallback
• Device Information - Shows GPU names for easy identification

Language Management

• 8 Language Support - Comprehensive language coverage
• Bidirectional Translation - Translate between any supported language pair
• Quick Swap Button (⇄) - Instantly reverse translation direction
• Smart Validation - Prevents selecting identical source/target languages

Text Processing

• Scrollable Text Areas - Handle long texts comfortably
• 5000 Character Limit - Optimized for quality and performance
• Clear Warnings - Helpful messages when text exceeds limits
• Copy/Clear Functions - Easy text management

Translation Optimization

• Advanced Prompting - Optimized prompts for better translation quality
• Smart Output Cleaning - Removes artifacts and explanatory text
• Tuned Parameters - Optimized generation settings for accuracy

📁 Model Structure

Your models/ directory should contain:

models/
├── config.json              # Model configuration
├── model.safetensors        # Model weights
├── tokenizer.json           # Tokenizer configuration
└── generation_config.json   # Generation parameters

🔧 Technical Features

Memory Optimization

• Efficient Loading - Uses accelerate library for optimized memory usage
• Mixed Precision - bfloat16 support to reduce VRAM consumption
• Smart Device Management - Automatic device allocation and detection

Error Handling

• Global Exception Handling - Comprehensive error catching
• Detailed Error Messages - Clear feedback for troubleshooting
• Automatic Logging - Errors saved to error_log.txt

User Experience

• Multi-threading - Non-blocking UI during translation
• Real-time Progress - Live status updates and progress indicators
• Responsive Design - Smooth interaction and feedback

🛠️ Troubleshooting

Common Issues

Model Loading Fails

• Verify all model files are present in models/ directory
• Check file permissions and integrity
• Ensure sufficient disk space

CUDA Errors

• Confirm PyTorch CUDA compatibility
• Update GPU drivers
• Check CUDA installation

Memory Issues

• Try CPU mode for large models
• Reduce input text length
• Close other GPU-intensive applications

Poor Translation Quality

• Verify model compatibility with selected language pair
• Check if model is specifically trained for translation
• Try shorter, simpler sentences

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📞 Support

If you encounter any issues or have questions, please open an issue on GitHub.