search

wonjune-kang / lvc-vc

End-to-End Zero-Shot Voice Conversion with Location-Variable Convolutions
https://arxiv.org/abs/2205.09784
MIT License
86 stars 6 forks source link

requirements.txt is incomplete #6

Closed dietmar closed 1 year ago

dietmar commented 1 year ago

Nice work, thank you for sharing!

I couldn't get the code to run with the provided requirements.txt file.

I got it to work with this requirements.txt on Python 3.10.12:

librosa==0.9.2
matplotlib==3.5.2
numpy>=1.22
scipy>=1.10.0
torch>=1.13.1
torchaudio==0.13.1
tensorboard==2.10.0
tqdm==4.64.0
omegaconf==2.2.3
praat-parselmouth
transformers

You might want to specify concrete versions for the last two. Cheers!

wonjune-kang commented 1 year ago

Thank you for pointing this out! I've updated requirements.txt to reflect your post and specified versions for praat-parselmouth and transformers, hope it solves the issues.