Menu ▾ ▴

Home

bendeb

🎵 Welcome to StemWeaver Wiki

StemWeaver v1.1 - Professional AI-Driven Audio Stem Separation Tool

Created by bendeb creations © 2026

📖 Table of Contents

Overview

StemWeaver is a powerful AI-driven audio stem separation tool that intelligently separates audio tracks into individual instrumental components. Using advanced Meta Demucs AI models, it can extract vocals, drums, bass, piano, guitar, and other instruments from any audio file.

Perfect for:

  • 🎹 Music producers
  • 🎧 DJs and remixers
  • 🎤 Content creators
  • 🔊 Audio professionals
  • 🎓 Music educators
  • 🎵 Audio researchers

Key Features

🤖 AI-Powered Technology

  • Meta Demucs Models: htdemucs_6s, htdemucs, mdx_extra, mdx_q
  • State-of-the-art audio separation algorithms
  • High-quality stem extraction

🎚️ 4 Extraction Profiles

  1. 6-Stem Mode: Vocals, Drums, Bass, Piano, Guitar, Other
  2. 4-Stem Mode: Vocals, Drums, Bass, Other
  3. Vocals Only: Extract vocals exclusively
  4. Instruments Only: Everything except vocals

📊 Quality Control

  • 24-bit audio preset for high quality
  • 32-bit audio preset for professional use
  • Customizable output formats
  • Metadata preservation

🖥️ Modern Interface

  • Beautiful, intuitive GUI built with DearPyGUI
  • Real-time processing feedback
  • Progress tracking
  • Drag-and-drop file support

💻 Hardware Acceleration

  • GPU Support: NVIDIA CUDA acceleration
  • CPU Fallback: Works without GPU
  • Automatic device detection
  • Optimized performance

📦 Distribution

  • Universal AppImage: Runs on any Linux distribution
  • Portable: No installation required
  • Self-contained: All dependencies included

Quick Start

For Users (AppImage)

  1. Download the latest AppImage from Releases

  2. Make it executable:
    bash chmod +x StemWeaver-v1.1-x86_64.AppImage

  3. Run StemWeaver:
    bash ./StemWeaver-v1.1-x86_64.AppImage

  4. Load your audio file and select extraction profile

  5. Click "Start Processing" and wait for results

That's it! Your separated stems will be in the output folder.

Installation

Pros: Works on any Linux distribution, no dependencies, portable
Cons: Larger file size

# Download
wget https://github.com/mangoban/StemWeaver/releases/download/v1.1/StemWeaver-v1.1-x86_64.AppImage

# Make executable
chmod +x StemWeaver-v1.1-x86_64.AppImage

# Run
./StemWeaver-v1.1-x86_64.AppImage

Method 2: Build from Source

Pros: Latest development version, customizable
Cons: Requires dependencies installation

# Clone repository
git clone https://github.com/mangoban/StemWeaver.git
cd StemWeaver

# Build AppImage
cd packaging
./build_appimage.sh

Method 3: Manual Installation (Advanced)

Pros: Direct control, development-friendly
Cons: Manual dependency management

On Manjaro/Arch Linux:

# Install system dependencies
sudo pacman -S python-pip ffmpeg

# Install Python packages
pip install dearpygui torch torchaudio demucs librosa soundfile pretty_midi midiutil

# Run from source
python gui_data/gui_modern_extractor.py

On Ubuntu/Debian:

# Install system dependencies
sudo apt install python3-pip ffmpeg

# Install Python packages
pip3 install dearpygui torch torchaudio demucs librosa soundfile pretty_midi midiutil

# Run from source
python3 gui_data/gui_modern_extractor.py

On Fedora:

# Install system dependencies
sudo dnf install python3-pip ffmpeg

# Install Python packages
pip3 install dearpygui torch torchaudio demucs librosa soundfile pretty_midi midiutil

# Run from source
python3 gui_data/gui_modern_extractor.py

System Requirements

Minimum Requirements

  • OS: Linux x86_64 (Any distribution)
  • RAM: 8GB
  • Storage: 400MB for app + models
  • CPU: Dual-core processor
  • Internet: For initial model download
  • RAM: 16GB+ (for 6-stem processing)
  • GPU: NVIDIA GPU with CUDA support
  • CPU: Quad-core or better
  • Storage: 1GB+ free space for output files

Supported Linux Distributions

  • ✅ Manjaro
  • ✅ Arch Linux
  • ✅ Ubuntu/Kubuntu/Xubuntu
  • ✅ Debian
  • ✅ Fedora
  • ✅ Linux Mint
  • ✅ openSUSE
  • ✅ Pop!_OS
  • ✅ Any modern Linux distribution

How to Use

Step 1: Launch StemWeaver

./StemWeaver-v1.1-x86_64.AppImage

Step 2: Select Input Audio File

  • Click "Browse" or drag and drop your audio file
  • Supported formats: MP3, WAV, FLAC, OGG, M4A, and more

Step 3: Choose Extraction Profile

  • 6-Stem: Full separation (Vocals, Drums, Bass, Piano, Guitar, Other)
  • 4-Stem: Standard separation (Vocals, Drums, Bass, Other)
  • Vocals Only: Extract only vocals
  • Instruments Only: Everything except vocals

Step 4: Select Quality Preset

  • 24-bit: High quality (recommended)
  • 32-bit: Professional quality (larger files)

Step 5: Choose Output Location

  • Default: ~/Music/StemWeaver_Output/
  • Or select custom folder

Step 6: Start Processing

  • Click "Start Processing"
  • Monitor progress in real-time
  • Processing time depends on profile and hardware

Step 7: Access Your Stems

  • Output folder structure:
    StemWeaver_Output/ └── [Song_Name]_[YYYYMMDD_HHMMSS]/ ├── vocals. wav ├── drums. wav ├── bass. wav ├── piano. wav (6-stem only) ├── guitar. wav (6-stem only) └── other.wav

Performance

Processing Times with GPU (NVIDIA CUDA)

Profile Time per Song RAM Usage
6-Stem 2-3 minutes 4-6 GB
4-Stem 1-2 minutes 3-4 GB
Vocals Only 1 minute 2-3 GB
Instruments Only 1 minute 2-3 GB

Processing Times with CPU Only

Profile Time per Song RAM Usage
6-Stem 5-10 minutes 6-8 GB
4-Stem 3-5 minutes 4-6 GB
Vocals Only 2-3 minutes 3-4 GB
Instruments Only 2-3 minutes 3-4 GB

Note: Times are approximate for a 3-4 minute song. Actual times vary based on hardware, song length, and complexity.

Use Cases

🎹 Music Production

  • Create acapellas for remixes
  • Extract instrumental tracks
  • Isolate specific instruments for sampling
  • Create stems for mixing

🎧 DJs & Remixers

  • Create custom mashups
  • Build transition tracks
  • Extract loops and samples
  • Prepare performance stems

🎤 Karaoke Creation

  • Remove vocals for backing tracks
  • Create instrumental versions
  • Extract vocals for learning

🔊 Audio Analysis

  • Study individual instrument performances
  • Educational purposes
  • Music transcription
  • Audio forensics

🎵 Content Creation

  • Create background music
  • Extract sound effects
  • Podcast production
  • Video production

🎓 Music Education

  • Teach instrument recognition
  • Study arrangement techniques
  • Analyze production methods
  • Practice mixing

Project Structure

StemWeaver/
├── README.md                    # Main documentation
├── LICENSE                      # CC BY 4.0 license
├── CONTRIBUTING.md              # Contribution guidelines
├── requirements.txt             # Python dependencies
├── Dockerfile                   # Docker container config
│
├── gui_data/                    # GUI application
│   ├── gui_modern_extractor.py # Main GUI application
│   ├── icons. py                 # Icon definitions
│   ├── ui_styling.py            # UI theming
│   └── fonts/                   # Font files
│
├── lib_v5/                      # AI model libraries
│   ├── mdxnet.py               # MDX-Net implementation
│   ├── tfc_tdf_v3.py           # Demucs components
│   └── vr_network/             # VR network models
│
├── models/                      # Pre-trained AI models
│   ├── Demucs_Models/          # Demucs models
│   └── MDX_Net_Models/         # MDX-Net models
│
├── packaging/                   # Distribution packages
│   ├── PKGBUILD                # Arch Linux package
│   └── build_appimage.sh       # AppImage builder
│
├── scripts/                     # Utility scripts
│   └── install_dependencies.sh
│
└── . github/                     # GitHub configuration
    └── workflows/               # CI/CD workflows

Development

Setting Up Development Environment

1. Clone the Repository

git clone https://github.com/mangoban/StemWeaver.git
cd StemWeaver

2. Create Virtual Environment

python -m venv venv
source venv/bin/activate  # On Linux/Mac

3. Install Dependencies

pip install -r requirements.txt

4. Install Additional Tools

# For AppImage building
sudo pacman -S appimagetool  # Arch/Manjaro
sudo apt install appimagetool  # Ubuntu/Debian

5. Run Development Version

python gui_data/gui_modern_extractor. py

Building the AppImage

cd packaging
./build_appimage. sh

This creates StemWeaver-v1.1-x86_64.AppImage in the root directory.

Building Arch Linux Package

cd packaging
makepkg -si

Code Style Guidelines

  • Python: Follow PEP 8
  • Comments: Clear and descriptive
  • Docstrings: Use for all functions
  • Type Hints: Use where appropriate
  • Testing: Write tests for new features

Testing

# Run unit tests
python -m pytest tests/

# Run GUI in test mode
python gui_data/gui_modern_extractor.py --test

Contributing

We welcome contributions from the community! Here's how you can help:

Ways to Contribute

  1. Report Bugs: Open an issue with detailed reproduction steps
  2. Suggest Features: Share your ideas for improvements
  3. Submit Code: Create pull requests with enhancements
  4. Improve Documentation: Help make docs clearer
  5. Test: Try new releases and report issues
  6. Translate: Help localize StemWeaver

Contribution Process

  1. Fork the repository
  2. Create a feature branch
    bash git checkout -b feature/AmazingFeature
  3. Make your changes
  4. Commit with clear messages
    bash git commit -m 'Add some AmazingFeature'
  5. Push to your fork
    bash git push origin feature/AmazingFeature
  6. Open a Pull Request

Code of Conduct

  • Be respectful and inclusive
  • Provide constructive feedback
  • Focus on what is best for the community
  • Show empathy towards others

For more details, see CONTRIBUTING.md (github.com)

Support & Resources

📚 Documentation

🐛 Issue Tracker

💬 Community

  • Discussions: GitHub Discussions
  • Developer: bendeb creations
  • Contact: contact@bendebcreations.com

☕ Support Development

License & Attribution

📜 License

StemWeaver v1.1 is licensed under Creative Commons Attribution 4.0 International (CC BY 4.0)

Copyright © 2026 bendeb creations

✅ What You CAN Do

  • ✓ Use for commercial purposes
  • ✓ Use for personal projects
  • ✓ Modify the software
  • ✓ Distribute to others
  • ✓ Create derivatives
  • ✓ Sell products made with StemWeaver

⚠️ What You MUST Do

  • ✓ Give credit to bendeb creations
  • ✓ Include the license
  • ✓ State changes if modified
  • ✓ Link back to the original project

📝 Proper Attribution

When using StemWeaver, include this notice:

StemWeaver v1.1 by bendeb creations
Licensed under Creative Commons Attribution 4.0 International (CC BY 4.0)
Original project: https://github.com/mangoban/StemWeaver

🙏 Credits

Developed by: bendeb creations
AI Technology: Meta AI Research (Demucs)
Framework: PyTorch Community
GUI: DearPyGUI

📄 Full License

For the complete license text, see LICENSE

Version History

v1.1.0 - January 6, 2026

  • ✅ AI-powered multi-stem audio separation
  • ✅ 4 extraction profiles (6-stem, 4-stem, vocals-only, instruments-only)
  • ✅ Modern professional GUI interface
  • ✅ GPU acceleration support (CUDA)
  • ✅ Universal Linux AppImage distribution
  • ✅ 24-bit and 32-bit quality presets
  • ✅ Real-time processing feedback
  • ✅ Comprehensive documentation
  • ✅ Creative Commons Attribution 4.0 licensing

FAQ

Q: Does StemWeaver work on Windows or macOS?

A: Currently, StemWeaver is Linux-only. Windows and macOS support may come in future versions.

Q: Do I need a GPU?

A: No, but GPU (NVIDIA CUDA) significantly speeds up processing.

Q: What audio formats are supported?

A: MP3, WAV, FLAC, OGG, M4A, and most common audio formats.

Q: Can I use StemWeaver commercially?

A: Yes! With proper attribution (CC BY 4.0 license).

Q: How accurate is the stem separation?

A: StemWeaver uses state-of-the-art Meta Demucs AI models, providing professional-quality separation.

Q: Where are the AI models stored?

A: Models are downloaded automatically on first use and cached in ~/.cache/torch/hub/demucs/

Q: Can I contribute to StemWeaver?

A: Absolutely! See the Contributing section.

StemWeaver v1.1
Professional Audio Stem Separation
By bendeb creations
© 2026 - Licensed under CC BY 4.0

🎵 Extract. Create. Inspire.

Back to Top

MongoDB Logo MongoDB