AI-powered voice cloning made simple - Optimized Edition

Record your voice for 10 seconds, then generate speech in your cloned voice with any text. Browser-based recording, no microphone setup required.

• 🎤 Browser Recording - No microphone setup, record directly in your browser
• 🤖 AI Voice Cloning - Powered by NeuTTS-Air (Qwen 0.5B backbone)
• 📝 Auto-Generated Prompts - Read a random sentence, we handle the rest
• 🎵 Text-to-Speech - Generate speech from any text in your voice
• 📱 Mobile Friendly - Works on phones, tablets, and desktops
• 🐳 Docker Ready - One-command deployment
• 💾 Persistent Storage - Your recordings and outputs are saved

# Clone the repository
git clone https://github.com/aldervall/clone-your-voice.git
cd clone-your-voice

# Start with Docker Compose
docker-compose up -d

# Open in browser
open http://localhost:5000

That's it! The interface is ready to use.

For more detailed setup options, including local development, see the Quick Start Guide.

Step 1: Record Your Voice (10 seconds)

• Click the microphone button
• Read the displayed prompt aloud
• Preview your recording

Step 2: Generate Speech

• Type any text you want to hear
• Click "Generate Speech"
• Wait for AI processing (~10-30 seconds)

Step 3: Download

• Listen to your generated audio
• Download the file
• Generate more!

• Quick Start Guide - Get running in 5 minutes
• Docker Deployment - Complete deployment guide
• GEMINI.md - Comprehensive context for AI agents (including project structure, technologies, and conventions)

• AI Model: NeuTTS-Air - Qwen 0.5B backbone
• Audio Codec: NeuCodec (50Hz neural codec)
• Backend: Python 3.11 + Flask
• Frontend: Vanilla JavaScript + CSS
• Deployment: Docker + Docker Compose

Contributions welcome! Fork, make changes, submit PR.

MIT License - see LICENSE

• NeuTTS-Air - Core TTS engine
• Neuphonic - AI model development

• GitHub: @aldervall
• Issues: Report here

Clone Your Voice - AI-powered voice cloning made simple 🎙️