VoiceCast

Your words, any voice.

Voice cloning and text-to-speech with multiple TTS engines. Clone any voice from a short audio sample and generate speech in that voice.

Features

• Voice Cloning - Clone voices from 5-30 second audio samples
• Multiple Engines - Coqui XTTS v2 (multilingual) and Chatterbox (fast/expressive)
• 16 Languages - English, Spanish, French, German, Chinese, Japanese, and more
• Three Interfaces - GUI application, CLI tool, and Python API
• Expressive Speech - Paralinguistic tags for laughs, sighs, gasps (Chatterbox Turbo)

Quick Start

Installation

# Clone repository
git clone https://github.com/luongnv89/voice-cast.git
cd voicecast

# Create virtual environment
python3.10 -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate

# Install dependencies
pip install -e .

# Optional: Chatterbox engine
pip install -e ".[chatterbox]"

Usage

GUI Application:

python voice_cloning_app.py

Command Line:

python vcloner.py -i voice.wav -t "Hello world" -o output.wav

Python API:

from voice_cloner import VoiceCloner

cloner = VoiceCloner(speaker_wav="./voice-samples/speaker.wav")
cloner.say("Hello, this is my cloned voice!", save_audio=True, output_file="output.wav")

TTS Engines

Engine	Languages	Speed	Best For
Coqui XTTS v2	16	Medium	Multilingual, quality
Chatterbox Turbo	English	Fast	Rapid iteration, expressions
Chatterbox Standard	English	Medium	Production quality

Expressive speech with Chatterbox Turbo:

cloner.say("That's hilarious [laugh]! I can't believe it [gasp]!")

Tags: [laugh], [chuckle], [cough], [sigh], [gasp], [yawn]

Documentation

Document	Description
API Reference	Complete Python API documentation
CLI Reference	Command-line interface guide
GUI Guide	Desktop application user manual
Engines Guide	TTS engine comparison and parameters
Architecture	System design and patterns
Development	Contributing and setup guide
Troubleshooting	Common issues and solutions

System Requirements

• Python 3.10+
• 8GB RAM (16GB recommended)
• NVIDIA GPU with CUDA (optional, for faster processing)

License

MIT License - see LICENSE file.

Acknowledgments

• Coqui TTS - XTTS v2 model
• Chatterbox - Fast TTS by Resemble AI
• PyTorch - Deep learning framework