Version of Hi-Fi GAN designed to work with
- Models can be exported to onnx format
- Python 3.7 or higher
- PyTorch 1.6 or higher
- librosa
git clone https://github.com/rhasspy/hifi-gan-train
cd hifi-gan-train
python3 -m venv .venv
source .venv/bin/activate
pip3 install --upgrade pip
pip3 install --upgrade wheel setuptool
pip3 install -r requirements.txt
bin/hifi-gan-train --debug /path/to/model --config /path/to/config.json < /path/to/wav_paths.txt
See the configs
directory for example configs and --help
for more options.