Voice Clone Engine

A robust, flexible voice cloning system with advanced speaker diarization and configurable resource management.

Features

• Advanced speaker diarization
• Flexible resource allocation (CPU/GPU)
• Comprehensive test coverage
• Efficient caching system
• Modular pipeline design

Requirements

• Python 3.13.2
• PyTorch 2.1.0
• torchaudio 2.1.0
• Compatible with both CPU and GPU environments

Installation

1. Clone the repository:

git clone [your-repo-url]
cd voice-clone-engine

2. Install dependencies:

pip install -r requirements.txt

Usage

🚧 This project is currently under development. For development status and planned features, see:

• REFACTOR_PLAN.md - Detailed development roadmap
• REQUIREMENTS.md - Project specifications and requirements

Resource Management

The system supports flexible resource allocation:

from src.resource_manager import ResourceManager

with ResourceManager() as rm:
    # Get available resource information
    resource_info = rm.get_resource_info()
    
    # Configure resource usage
    rm.configure("--resource all")                # Use all available resources
    rm.configure("--resource cpu")                # CPU only
    rm.configure("--resource gpu")                # GPU only
    rm.configure("--resource count:3")            # Use any 3 available resources
    rm.configure("--resource-exclude cpu")        # Use all except CPU
    rm.configure("--resource-memory-limit 80")    # Limit memory usage to 80%
    rm.configure("--resource-strategy balanced")  # Distribute load evenly

Running Demos

The project includes several demos:

cd demos
python embedding_demo.py     # Voice embedding visualization
python diarization_demo.py   # Speaker diarization
python caching_demo.py       # Performance impact of caching

Project Structure

voice-clone-engine/
├── configs/           # Configuration files
├── data/             # Dataset storage (gitignored)
├── demos/            # Demo scripts
├── src/              # Source code
│   ├── speaker_diarizer.py
│   └── resource_manager.py
├── tests/            # Unit tests
└── plots/            # Visualization outputs

Testing

Run the test suite:

python -m pytest tests/

For development guidelines and testing practices, see REFACTOR_PLAN.md.

License

[Your chosen license]

Contributing

See REFACTOR_PLAN.md for development guidelines and workflow.