Stemify Desktop is a professional desktop application for AI-powered audio stem separation. Built with Facebook Research's Demucs model, it allows users to separate audio tracks into individual components (vocals, drums, bass, other) with high accuracy.

✨ Features

🖥️ Desktop Application - Native app for Windows, macOS, and Linux
🤖 AI-Powered Separation - Uses state-of-the-art Demucs AI model
⚡ Real-time Processing - Fast audio separation with progress tracking
🎵 Multiple Formats - Support for MP3 files up to 20MB
🎨 Professional UI - Modern interface with custom branding
📱 Offline Capability - Works without internet connection
🔄 Drag & Drop - Intuitive file upload interface

🚀 Downloads

Windows

Stemify-1.0.0-win-universal.zip (93.1 MB) - Universal installer (x64 + x86 + ARM64)

macOS

Stemify-1.0.0-macos-intel.dmg (119 MB) - Intel Mac (x64)
Stemify-1.0.0-macos-arm64.dmg (114 MB) - Apple Silicon (ARM64)

Linux

Stemify-1.0.0.AppImage (120 MB) - Intel/AMD (x64)
Stemify-1.0.0-arm64.AppImage (121 MB) - ARM64

🎯 How to Use

Download the appropriate file for your operating system
Install the application following your OS instructions
Launch Stemify Desktop
Upload your audio file (drag & drop or click)
Wait for AI processing
Download your separated tracks

🛠️ Technical Details

Architecture

Frontend: React + TypeScript + Vite
Backend: Flask + Python
Desktop: Electron wrapper
AI Model: Facebook Research Demucs
Audio Processing: Real-time stem separation

Supported Formats

Input: MP3 files up to 20MB
Output: 4 stems (Vocals, Drums, Bass, Other)
Quality: High-fidelity separation

System Requirements

Windows: Windows 10+ (ARM64)
macOS: macOS 10.12+ (Apple Silicon)
Linux: Modern Linux distribution (ARM64)
Memory: 4GB RAM minimum
Storage: 500MB free space

🧠 Model Information

Stemify uses the Demucs v4 model from Facebook Research, specifically trained for high-quality music source separation. The model has been trained on large datasets and can separate:

🎤 Vocals - Lead and backing vocals
🥁 Drums - Complete drum kit
🎸 Bass - Bass guitar and low-frequency elements
🎹 Other - Remaining instruments (guitar, piano, synths, etc.)

📊 Performance

Processing Time: 2-5 minutes for typical 3-5 minute tracks
Quality: Professional-grade separation
Memory Usage: ~2GB during processing
CPU Usage: Optimized for ARM64 processors

🔧 Build from Source

bash

git clone https://github.com/huchukato/stemify-audio-splitter.git
cd stemify-audio-splitter/demucs-gui
npm install
npm run electron-build -- --mac --win --linux

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.

📄 License

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

🙏 Acknowledgments

Facebook Research for the Demucs model
Electron Team for the desktop framework
React Community for the frontend library
Python Audio Community for audio processing tools

� Author

huchukato
- 🐙 GitHub
- 🐦 X (Twitter)
- 🎨 Civitai - Check out my AI art models!

�📞 Support

GitHub Issues: Report bugs
Discussions: Community forum
Email: Contact through GitHub profile

Made with ❤️ by huchukato

Transform your music with AI-powered stem separation!

huchukato/stemify-desktop

🎵 Stemify Desktop - The Audio Splitter

📝 Description