Hugging Face
Models
Datasets
Spaces
Posts
Docs
Enterprise
Pricing
Log In
Sign Up
Datasets:
jiaheillu
/
sovits_audio_preview
like
7
Modalities:
Audio
Image
Formats:
soundfolder
Languages:
Afar
Size:
< 1K
Tags:
music
Libraries:
Datasets
Croissant
License:
openrail
Dataset card
Viewer
Files
Files and versions
Community
3
f75b822
sovits_audio_preview
/
README.md
jiaheillu
Update README.md
f75b822
almost 2 years ago
preview
code
|
raw
Copy download link
history
blame
Safe
319 Bytes
# 效果预览
本仓库用于预览训练出的各种语音模型的效果
各文件主要内容:
关键参数:
音频时长:min
epoch: 轮
无关参数:
batch
_size =
segments =
step=segments*epoch/batch_
size,即模型文件后面数字由来
损失函数图像:主要看step 与 loss5