Multiple state of the art neural audio synthesis models refactored to share a uniform interface.
The goal of this project is to make it easy to train and run a variety of different neural synthesis models from the command line.
Create an anaconda environment to gather dependencies for this repository:
conda create -n mauaudio python=3.8 pytorch torchvision torchaudio cudatoolkit=11.3 cudatoolkit-dev=11.3 pip=21.3.1 mpi4py Cython -c pytorch -c nvidia -c conda-forge
conda activate mauaudio
pip install -r requirements.txt