SUImodels / README.md
Miuzarte's picture
Update README.md
ebc191c
|
raw
history blame
5.64 kB
---
language:
- zh
tags:
- AIvtuber
- VirtuaReal
---
# SUImodels
### 岁己所有的模型都在这里
### 包括sovits3.0、4.0及onnx,还有以后会出的vits模型
## Model Description
<!-- Provide a longer summary of what this model is. -->
- **Developed by:** [謬紗特](https://space.bilibili.com/59442895)
- **Model type:** [so-vits-svc 3.0 48kHz](https://github.com/innnky/so-vits-svc/tree/main)、[so-vits-svc 4.0](https://github.com/innnky/so-vits-svc/tree/4.0)
- **Demo:** [SPACE: AI岁己(歌声变声器)](https://huggingface.co/spaces/Miuzarte/SUI-svc-3.0)
### pth文件名的训练步数是程序按学习率等超参数得出的步数,onnx文件名的步数为实际训练步数
|Sovits3_v1|Base/G_1000000.pth|Singing/G_1M111000.pth|Singing/G_100000.pth|
|-:|:-:|:-:|:-:|
|onnx|Base/suijiSUI_v1_1M_SoVits.onnx|Singing/suijiSUI_v1_1M111000_SoVits.onnx|Singing/suijiSUI_v1_100000_SoVits.onnx|
|训练集|12月录播(除电台)、出道至今22条歌投、10条歌切、圣诞音声(27.5小时)|Base/G_1000000.pth作为底模_2022年所有唱歌投稿、唱歌切片、圣诞音声(3.9小时)|2022年所有唱歌投稿、唱歌切片、圣诞音声(3.9小时)|
### 因为v2练着练着突然sovits4.0就出来了所以200k直接收了,弃用
|Sovits3_v2|Base/G_100000.pth|Singing/G_160000.pth|
|-:|:-:|:-:|
|onnx|Base/suijiSUI_v2_100000_SoVits.onnx|Singing/suijiSUI_v2_100k100000_SoVits.onnx|
|训练集|22年12月、23年1月的录播(06:47:46)|Base/G_100000.pth作为底模_22年12月、23年1月、23年2月1-17日的录播(除电台,共计268:07:43)、岁己的投稿、A1in_sy11月及以前的歌切|
### Singing/G_160000.pth、Singing/G_210000.pth、Singing/G_260000.pth我个人倾向于折中使用Singing/G_210000.pth
|Sovits4_v3|Base/G_100000.pth|Singing/G_160000.pth|
|-:|:-:|:-:|
|onnx|Base/suijiSUI_v3_100000_SoVits.onnx|Singing/suijiSUI_v3_100k100000_SoVits.onnx|
|训练集|22年12月、23年1月的录播(06:47:46)|Base/G_100000.pth作为底模_22年12月、23年1月、23年2月1-17日的录播(除电台,共计268:07:43)、岁己的投稿、A1in_sy11月及以前的歌切|
|Sovits4_v3|Singing/G_210000.pth|Singing/G_260000.pth|备注|
|-:|:-:|:-:|:-:|
|onnx|Singing/suijiSUI_v3_100k150000_SoVits.onnx|Singing/suijiSUI_v3_100k200000_SoVits.onnx|160k开始loss就没再往下了,这两个估计有一丁点过拟合,然后我个人|
|训练集|{同Singing/G_160000.pth}|{同Singing/G_160000.pth}|也听不出这三个有什么区别,有强迫症的可以自己再仔细对比一下|
## MoeSS\\Mods配置文件
岁己SUI_v1_1M.json (suijiSUI_v1_1M\\)
```json
{
"Folder" : "suijiSUI_v1_1M",
"Name" : "岁己SUI_v1_1M",
"Type" : "SoVits",
"Rate" : 48000,
"Hop" : 320,
"Hubert": "hubert",
"SoVits3": true,
"Characters" : ["岁己SUI"]
}
```
岁己SUI_v1_1M111k.json (suijiSUI_v1_1M111000\\)
```json
{
"Folder" : "suijiSUI_v1_1M111000",
"Name" : "岁己SUI_v1_1M111k",
"Type" : "SoVits",
"Rate" : 48000,
"Hop" : 320,
"Hubert": "hubert",
"SoVits3": true,
"Characters" : ["岁己SUI"]
}
```
岁己SUI_v1_100k.json (suijiSUI_v1_1M\\)
```json
{
"Folder" : "suijiSUI_v1_1M",
"Name" : "岁己SUI_v1_1M",
"Type" : "SoVits",
"Rate" : 48000,
"Hop" : 320,
"Hubert": "hubert",
"SoVits3": true,
"Characters" : ["岁己SUI"]
}
```
岁己SUI_v2_100k.json (suijiSUI_v2_100000\\)
```json
{
"Folder" : "suijiSUI_v2_100000",
"Name" : "岁己SUI_v2_100k",
"Type" : "SoVits",
"Rate" : 48000,
"Hop" : 320,
"Hubert": "hubert",
"SoVits3": true,
"Characters" : ["岁己SUI"]
}
```
岁己SUI_v2_100k100k.json (suijiSUI_v2_100k100000\\)
```json
{
"Folder" : "suijiSUI_v2_100k100000",
"Name" : "岁己SUI_v2_100k100k",
"Type" : "SoVits",
"Rate" : 48000,
"Hop" : 320,
"Hubert": "hubert",
"SoVits3": true,
"Characters" : ["岁己SUI"]
}
```
岁己SUI_v3_100k.json (suijiSUI_v3_100000\\)
```json
{
"Folder" : "suijiSUI_v3_100000",
"Name" : "岁己SUI_v3_100k",
"Type" : "SoVits",
"Rate" : 44100,
"Hop" : 512,
"Hubert": "hubert4.0",
"SoVits4": true,
"Characters" : ["岁己SUI"]
}
```
岁己SUI_v3_100k100k.json (suijiSUI_v3_100k100000\\)
```json
{
"Folder" : "suijiSUI_v3_100k100000",
"Name" : "岁己SUI_v3_100k100k",
"Type" : "SoVits",
"Rate" : 44100,
"Hop" : 512,
"Hubert": "hubert4.0",
"SoVits4": true,
"Characters" : ["岁己SUI"]
}
```
岁己SUI_v3_100k150k.json (suijiSUI_v3_100k150000\\)
```json
{
"Folder" : "suijiSUI_v3_100k150000",
"Name" : "岁己SUI_v3_100k150k",
"Type" : "SoVits",
"Rate" : 44100,
"Hop" : 512,
"Hubert": "hubert4.0",
"SoVits4": true,
"Characters" : ["岁己SUI"]
}
```
岁己SUI_v3_100k200k.json (suijiSUI_v3_100k200000\\)
```json
{
"Folder" : "suijiSUI_v3_100k200000",
"Name" : "岁己SUI_v3_100k200k",
"Type" : "SoVits",
"Rate" : 44100,
"Hop" : 512,
"Hubert": "hubert4.0",
"SoVits4": true,
"Characters" : ["岁己SUI"]
}
```
# Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
## Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
## Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
## Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]