huggingface_hub==0.22.2
numpy
scipy
torch
librosa
pysptk
soundfile
wavenet_vocoder==0.1.1