search

yl4579 / StarGANv2-VC

StarGANv2-VC: A Diverse, Unsupervised, Non-parallel Framework for Natural-Sounding Voice Conversion
MIT License
466 stars 110 forks source link

How to train it for my own custom dataset? #93

Closed Hammad-Bangash closed 8 months ago

Hammad-Bangash commented 8 months ago

Actually, I want to train it for my own custom dataset. But it shows problems and does not get trained. Could you please specify what should be the audio properties for my own dataset so that i can train it. image this is the problem it shows while training: image

yl4579 commented 8 months ago

This is likely caused by wrong num_domain here: https://github.com/yl4579/StarGANv2-VC/blob/main/Configs/config.yml#L28 (and also your speaker id in the dataset).