刻晴_preview/README.md

# 刻晴语音模型训练参数



duration of dataset：50min 

epoch: 1822

batch_size :24

segments = 336

step:25500

Losses vs step

![截屏2023-04-08 13.20.22](./README.assets/%E6%88%AA%E5%B1%8F2023-04-08%2013.20.22.png)