Datasets:

jiaheillu
/

sovits_audio_preview

# 可莉语音模型训练参数



duration of dataset：39min 

epoch: 2819

batch_size :24

segments = 188

step:22400

Losses vs step

![截屏2023-04-08 21.51.13](./README.assets/%E6%88%AA%E5%B1%8F2023-04-08%2021.51.13.png)