DOCSEARCHbyAlgolia
Submit Docs

Ultimate Vocal Remover (UVR) GUI

repository·master·Indexed 12 days ago

https://github.com/anjok07/ultimatevocalremovergui

A source separation tool using deep learning models to remove vocals from audio files. It provides a graphical interface for stem separation, time-stretching, and pitch-shifting across Windows, macOS, and Linux. Supports Nvidia GPU acceleration via CUDA and AMD/Intel acceleration via DirectML.

Tokens
1.5K
Snippets
5
Records
8
Agent score
48%

What's inside Ultimate Vocal Remover

  1. License and usage requirements

    master

    The Ultimate Vocal Remover GUI code is licensed under the MIT License.

    Important for Developers: If you are a third-party application developer intending to use UVR models in your own software, you must honor the MIT license by providing proper credit to UVR and its developers.

  2. Hardware Requirements and Performance

    master

    To ensure stable and efficient performance, note the following:

    • GPU Requirements:
      • Minimum: Nvidia GTX 1060 6GB for GPU conversions.
      • Recommended: Nvidia GPU with at least 8GB V-RAM.
      • AMD Radeon support is currently limited (see specific AMD working branches).
    • Platform: Only compatible with 64-bit platforms.
    • Dependencies:
      • FFmpeg is required for non-WAV audio files.
      • Rubber Band is required for Time-Stretch and Pitch-Shift tools.
    • Settings: The application automatically remembers your settings upon closing.

  3. Install UVR on Linux (Debian & Arch)

    master

    Follow these steps for Debian-based (Ubuntu, Mint) or Arch-based (EndeavourOS) systems. It is highly recommended to use a virtual environment (venv) to avoid interfering with system Python packages.

    1. Install System Dependencies: Use apt for Debian or pacman for Arch to install ffmpeg, python3-pip, and tk.
    2. Setup Virtual Environment:
      • Navigate to the repo directory.
      • Create a venv: python3 -m venv venv.
      • Activate it: source venv/bin/activate.
    3. Install Python Dependencies: pip install -r requirements.txt.
    4. Run: python UVR.py.
    # Debian-based
sudo apt update && sudo apt upgrade
sudo apt-get install -y ffmpeg python3-pip python3-tk

# Arch-based
sudo pacman -Syu
sudo pacman -S ffmpeg python-pip tk

# Virtual Environment Setup
cd /path/to/ultimatevocalremovergui
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python UVR.py

  4. Install UVR on Windows

    master

    For Windows 10 or higher, use the official installer.

    Important Requirements:

    • Install UVR to the main C:\ drive to avoid instability.
    • Intel Pentium & Celeron CPUs and Windows 7 or lower are not guaranteed to work.

    Installer Options:

    • Standard Installer: For most users.
    • DirectML Version: Use this if you have an AMD Radeon or Intel Arc graphics card.
    • Patch: If you already have UVR installed, you can use the provided patch to update to v5.6.
    # Download links are provided in the README documentation for the .exe installers.

  5. Manual Windows Installation

    master

    If you prefer a manual setup, follow these steps:

    1. Download and extract the repository.
    2. Install Python 3.9.8 (ensure you check "Add python.exe to PATH" during installation).
    3. Install dependencies using pip.
    4. (Optional) If using an Nvidia GPU, upgrade torch with CUDA support.
    5. (Optional) Manually add FFmpeg and Rubber Band binaries to the UVR application directory to enable non-WAV processing and Time Stretch/Pitch Shift tools.
    # Install requirements
python.exe -m pip install -r requirements.txt

# For Nvidia GPU users
python.exe -m pip install --upgrade torch --extra-index-url https://download.pytorch.org/whl/cu117

  6. Install UVR on MacOS

    master

    UVR supports macOS Big Sur and above. M1 (arm64) users benefit from expanded MPS (GPU) acceleration for Demucs v4 and MDX-Net models.

    Installation Steps:

    1. Download the appropriate .dmg (arm64 for M1, x86_64 for Intel).
    2. Install Python 3.10.
    3. Install dependencies via pip3.
    4. (M1 Only) Run the specific command to fix libsndfile paths.
    5. (Optional) Manually add FFmpeg and Rubber Band binaries to the application directory for full functionality.
    # Install dependencies
pip3 install -r requirements.txt

# M1 Mac specific fix (run if on M1)
cp /Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/_soundfile_data/libsndfile_arm64.dylib /Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/_soundfile_data/libsndfile.dylib

  7. Troubleshoot MacOS App Security (Notarization)

    master

    If you cannot open the UVR application on MacOS due to Apple's security restrictions, run these commands in Terminal.app:

    1. Disable gatekeeper temporarily: sudo spctl --master-disable
    2. Bypass quarantine for the UVR app: sudo xattr -rd com.apple.quarantine /Applications/Ultimate\ Vocal\ Remover.app

    Note: It is recommended to re-enable gatekeeper after the app opens properly.

    sudo spctl --master-disable
sudo xattr -rd com.apple.quarantine /Applications/Ultimate\ Vocal\ Remover.app

  8. Troubleshoot Memory and FFmpeg Errors

    master

    Common issues and their resolutions:

    • Non-WAV conversion error: This occurs if FFmpeg is not installed or not found in the application directory.
    • Memory allocation errors: If you encounter memory issues during processing, try lowering the "Segment" or "Window" sizes in the application settings.
    • Reporting Issues: When reporting a bug, click the "Settings Button" (left of "Start Processing") and then click "Error Log" to provide detailed diagnostic information.