🎙 Universal Voice Changer CLI (Academic Edition)

A modular command-line interface for RVC (Retrieval-based Voice Conversion) and GSS engines. Optimized for performance and cross-platform compatibility (macOS ARM64, Windows, Linux).

---

🛠 Prerequisites

1. Environment Setup (Python 3.10)

Python 3.10 is strictly required. Use Conda for a stable setup:

conda create -n hrvc_env python=3.10 -y
conda activate hrvc_env

2. Dependency Installation

Always use the python -m pip prefix to ensure packages are installed in the correct environment.

For macOS (Apple Silicon ARM64)

python -m pip install "pip<24.1" "setuptools<70.0.0"
python -m pip install -r requirements.txt

For Windows / Linux (NVIDIA GPU)

# 🚨 CRITICAL: Run this FIRST to enable GPU (CUDA) acceleration:
python -m pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

# Then install the rest:
python -m pip install "pip<24.1" "setuptools<70.0.0"
python -m pip install -r requirements.txt

---

🐳 Running with Docker (Optional)

This project is fully containerized for reproducibility and server-side testing.

# Build the image and start the container in background
docker-compose up -d --build

# Enter the container environment
docker exec -it voice_changer_cli bash

# Execute voice conversion inside Docker (using generic paths)
docker exec -it voice_changer_cli python voice_cli.py -adapt hrvc_adapter -engine "./engine" -model "./model"

Warning

GPU acceleration and interactive Audio I/O might be limited inside Docker on macOS. Use for environment testing and server-side inference.

---

🚀 Usage

Run the CLI using your specific adapter and model paths:

python voice_cli.py -adapt hrvc_adapter -engine "./engine" -model "./model"

Command Arguments:

• -adapt: Engine adapter (e.g., hrvc_adapter).
• -engine: Path to the engine source folder.
• -model: Path to the voice model folder.
• --pitch: Semitone shift (-12 to 12).
• --index: Feature retrieval ratio (0.0 to 1.0).
• --protect: Voiceless protection threshold.

---

🏗 Project Architecture

• voice_cli.py: Interactive interface with real-time dB monitoring.
• adapters/: Modular engine adapters (decoupled logic).
• last_devices.json: Auto-saving your audio I/O configuration.

📄 License

MIT License. Optimized for Research and Academic Submission. 🏛