umap-learn visdom librosa matplotlib>=3.3.0 numpy scipy>=1.0.0 tqdm sounddevice SoundFile Unidecode inflect PyQt5 multiprocess numba webrtcvad pypinyin flask flask_wtf flask_cors gevent flask_restx tensorboard streamlit PyYAML torch_complex espnet PyWavelets monotonic-align==0.0.3 transformers fastapi loguru typer[all] click