Update README_zh_CN.md for whisper-ppg

This commit is contained in:
KakaruHayate 2023-05-25 21:41:50 +08:00 committed by GitHub
parent 42beea9a3f
commit 105c1c4197
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 7 additions and 2 deletions

View File

@ -86,8 +86,12 @@ wget -P pretrain/ http://obs.cstcloud.cn/share/obs/sankagenkeshi/checkpoint_best
##### **3. 若使用OnnxHubert/ContentVec作为声音编码器**
- download model at https://huggingface.co/NaruseMioShirakana/MoeSS-SUBModel/tree/main
- Place it under the `pretrain` directory
- 放在`pretrain`目录下
##### **4. 若使用Whisper-ppg作为声音编码器**
- download model at https://openaipublic.azureedge.net/main/whisper/models/345ae4da62f9b3d59415adc60127b97c714f32e89e936602e85993674d08dcb1/medium.pt
- 放在`pretrain`目录下
#### **编码器列表**
- "vec768l12"
- "vec256l9"
@ -97,6 +101,7 @@ wget -P pretrain/ http://obs.cstcloud.cn/share/obs/sankagenkeshi/checkpoint_best
- "vec768l12-onnx"
- "hubertsoft-onnx"
- "hubertsoft"
- "whisper-ppg"
#### **可选项(强烈建议使用)**
@ -174,7 +179,7 @@ python resample.py
### 2. 自动划分训练集、验证集,以及自动生成配置文件
```shell
python preprocess_flist_config.py --speech_encoder vec768l12
python preprocess_flist_config.py --speech_encoder whisper-ppg
```
speech_encoder拥有三个选择