# 鹿野院平藏语音模型训练参数



duration of dataset：67

epoch: 2223

batch_size :24

segments = 433

step:39800

Losses vs step

![截屏2023-04-16 11.39.01](./README.assets/%E6%88%AA%E5%B1%8F2023-04-16%2011.39.01.png)