Installation

Manual Installation

Windows

1. Download the latest precompiled .zip file from Hugging Face.
2. Extract the archive to a simple path on your C: drive (e.g., C:\Applio).
3. Run the run-applio.bat file to start Applio.

Important

• Make sure the installation path does not contain spaces or special characters.
• Temporarily disable your antivirus or firewall if it interferes with the installation.
• Do not run the installer as an administrator.

Linux/macOS

Option 1: Git Clone

1. Clone the repository from GitHub:

   git clone https://github.com/IAHispano/Applio.git
   cd Applio

2. Run the installation script:

   chmod +x run-install.sh
   ./run-install.sh

3. Launch Applio:

   chmod +x run-applio.sh
   ./run-applio.sh

Option 2: Manual Python Installation

1. Clone the repository from GitHub:

   git clone https://github.com/IAHispano/Applio.git
   cd Applio

2. Set up a Python virtual environment and install dependencies manually. See the Detailed Setup for Linux/macOS section below for instructions.

First-Time Setup

If you encounter issues, you may need to install Python 3.10.12 and create a virtual environment. See the Detailed Setup for Linux/macOS section below for instructions.

---

Advanced Installation & Guides

This section contains advanced installation methods and guides for specific hardware.

AMD GPU Support (Windows)

If you have an AMD GPU, you can set up Applio with Zluda for hardware acceleration.

1. Install Dependencies

1. Download and install the VC++ Runtime.
2. Verify your GPU meets the System Requirements.
3. Install the appropriate HIP SDK version from the AMD ROCm Hub for your GPU.

2. Install Applio

1. Download and extract the Applio v3.2.5+ precompiled version.
2. Open a command prompt in the Applio folder and run the following commands to install the correct PyTorch version:

   env\python -m pip uninstall torch torchvision torchaudio
   env\python -m pip install torch==2.3.1 torchvision==0.18.1 torchaudio==2.3.1 --upgrade --index-url https://download.pytorch.org/whl/cu118

3. Configure Zluda

1. Move all .bat files from the assets\zluda directory to the root Applio folder.
2. Run the appropriate patch script for your HIP SDK version (patch_zluda_hip57.bat or patch_zluda_hip61.bat).
3. Add the bin directory of your HIP SDK installation to your system’s Path environment variable.
4. Run run-applio-amd.bat to start Applio.

Initial Compilation

The first time you run Applio with Zluda, it will compile kernel code, which can take 15-20 minutes. The application may appear frozen during this time.

Detailed Setup for Linux/macOS

If the manual installation for Linux/macOS fails, you may need to set up a virtual environment manually.

Debian/Ubuntu

1. Ensure you have Python 3.10 installed.
2. Open a terminal in the Applio directory and run:

   sudo apt install python3.10-venv -y
   python3.10 -m venv .venv
   source .venv/bin/activate
   # Continue with installation steps inside the virtual environment

Arch Linux

1. Ensure you have Python 3.10 installed.
2. Open a terminal in the Applio directory and run:

   sudo pacman -S python-virtualenv --noconfirm
   python3.10 -m venv .venv
   source .venv/bin/activate
   # Continue with installation steps inside the virtual environment

Docker Deployment

Applio provides official Docker images for both CPU and GPU (CUDA) deployments.

1. Install Docker: Follow the official Docker installation guide for your operating system. For GPU support, install NVIDIA Container Toolkit.
2. Clone the repository:

   git clone https://github.com/IAHispano/Applio.git
   cd Applio

3. Build the Docker image:
   • For CPU: docker build -f Dockerfile -t applio:cpu .
   • For GPU: docker build -f Dockerfile_cuda -t applio:cuda .
4. Run the container:
   • Using Docker Compose: docker compose up
   • Or manually: docker run --gpus all -p 6969:6969 applio:cuda
5. Access Applio: Open http://localhost:6969 in your browser.

Tip

Docker is an excellent option for server deployments, cloud environments, and ensuring consistent dependencies across different systems.

---

Updating and Uninstalling

To update Applio, we recommend deleting the current Applio folder and reinstalling using your preferred method, but be sure to back up any desired data first! Here are the most common desired data locations:

• assets/audios: Audio you have uploaded, and the inference results.
• logs: Inference models you have downloaded or trained, plus some other stuff that comes with Applio.
• rvc/models/pretraineds/custom: Any pretrained base models you have downloaded.

1. Move or copy the above folders out of the Applio folder, if they contain something you want to keep.
2. Delete the entire Applio folder, then install the new version.
3. Move the above folders back into their correct locations in the new Applio folder. Merge folders with the same names, but do not overwrite files that came with both Applio versions. This should mean choosing to “merge” or “write into” folders with the same names, but choosing to “skip” files with the same names.

To uninstall Applio, simply delete the Applio folder.