- Python v == 3.6
- Numpy v == 1.19.5
- torch v == 1.10.1
- wandb v == 0.12.11
- SoundFile v ==0.10.2
- librosa v ==0.9.1
python3 train.py --urban_noise False (speech enhancement experiments on the DAPS dataset)
python3 train.py --urban_noise True (speech enhancement experiments on the BBC dataset)