1. Installation

• Using Conda
• Using pip

Using Conda

The easiest way to use Spleeter is probably to use it with conda. You need first to install conda, for instance using miniconda.

We provide two conda environment files :

• spleeter-cpu.yaml for CPU devices.
• spleeter-gpu.yaml for GPU devices¹.

Assuming that you have a git client installed, you can create and activate your environment as following :

git clone https://github.com/deezer/spleeter
conda env create -f spleeter/conda/spleeter-cpu.yaml
conda activate spleeter-cpu

All dependencies required to make Spleeter work would be installed in your environment as well as Spleeter library itself.

¹ running Spleeter on Nvidia GPU devices require you to have a proper CUDA driver installation.

Using pip

Spleeter could also be installed directly using pip package manager, assuming that following dependencies are installed first :

• ffmpeg
• libsndfile²

² only needed if you plan to perform evaluation.

If you do not feel confortable installing these dependencies, we recommend an installation with conda.

You can then install the library from PyPi repository :

pip install spleeter

⚠️ pip based install method does not support GPU. If you need to run Spleeter on GPU devices, we recommend an installation with conda or to use the provided GPU Docker image.