gradio
librosa
numpy
soundfile
transformers
pyannote.audio
torchaudio
torch