This work is the implemention of SiD-Waveflow.

Visit our website for audio samples.

1. Clone our repo and initialize submodule

   git clone https://github.com/NVIDIA/waveglow.git
   cd waveglow
   git submodule init
   git submodule update

2. Install requirements pip3 install -r requirements.txt

3. Install Apex

1. Download CSMSC. In this example it's in ~/BBdata/

2. Train

   mkdir checkpoints
   python train.py -c config.json

   For mixed precision training set "fp16_run": true on config.json.

3. Make test set mel-spectrograms

   python mel2samp.py -f traintestset_chn/test_files_copy.txt -o ./inferaudio/chn_mel -c config.json

4. Do inference with your network

   ls inferaudio/chn_mel/*.pt > mel_files.txt
   python3 inference.py -f mel_files.txt -w checkpoints/test1_chn_model -o ./inferaudio --is_fp16 -s 0.6