Voice Conversion Coming!

[huangxu1991]

Without Any Retrain! Plz ref. https://github.com/huangxu1991/GPT-SoVITS-VC. All stay the same except python vc_webui.py

Edited: Chinese demos: https://github.com/huangxu1991/GPT-SoVITS-VC English demos: https://github.com/huangxu1991/GPT-SoVITS-VC/blob/main/docs/en/README.md

[huangxu1991]

声音转换/变声/VC 来了 (项目地址: https://github.com/huangxu1991/GPT-SoVITS-VC) 请先按照源项目配置完环境，然后执行 python vc_webui.py

UI 上传以下这些：

1. 你想要的目标音色
2. 你想要转换的源音频
3. 源音频对应文本
4. 文本对应语言 一键合成即可。

新增 demo 试听如下： 中文 demos: https://github.com/huangxu1991/GPT-SoVITS-VC 英文 demos: https://github.com/huangxu1991/GPT-SoVITS-VC/blob/main/docs/en/README.md

[KakaruHayate]

emmmmm...非常有创意的工作...

[huangxu1991]

emmmmm...非常有创意的工作...

😜

[invokermoon]

致敬

[k18coldwl]

interesting!

[BankNatchapol]

Thanks for your amazing work, @huangxu1991 ! I've tested it, and it functions superbly. However, I face an issue – I prefer processing audio directly without adding corresponding text, but this will result in some word misspelling. Any suggestions on how to address this? Appreciate your assistance in advance!

[huangxu1991]

Thanks for your amazing work, @huangxu1991 ! I've tested it, and it functions superbly. However, I face an issue – I prefer processing audio directly without adding corresponding text, but this will result in some word misspelling. Any suggestions on how to address this? Appreciate your assistance in advance!

Good question! It happens because semantic ids extracted from RVQ lose text infomation more or less! So explictly providing parallel TEXT as info supplements!

[RVC-Boss]

interesting

[tianwang1122]

无需任何重新训练！请参考。https://github.com/huangxu1991/GPT-SoVITS-VC。 一切都保持不变，除了python vc_webui.py

编辑： 中文演示： https: //github.com/huangxu1991/GPT-SoVITS-VC 英文演示：https://github.com/huangxu1991/GPT-SoVITS-VC/blob/main/docs/en/README.md

为什么我的无法运行呢，我用的是整合包GPT-SoVITS-beta0217下载好您的整合包之后点击python vc_webui.py弹出命令窗口然后立马关闭

[tianwang1122]

消耗任何重新训练！请参考。https://github.com/huangxu1991/GPT-SoVITS-VC。一切都保持不变，除了python vc_webui.py 编辑： 中文演示： https://github.com/huangxu1991/GPT-SoVITS-VC中文演示：https://github.com/huangxu1991/GPT-SoVITS-VC/blob/main/docs/en/README.md

我的无法运行呢，我用的是集成包GPT-SoVITS-beta0217下载好您的集成包之后点击为什么python vc_webui.py弹出命令窗口然后立马关闭

https://github.com/RVC-Boss/GPT-SoVITS/assets/152373957/485c88cf-9ea6-462e-9c20-2ac98094948a 这是运行情况

[huangxu1991]

消耗任何重新训练！请参考。https://github.com/huangxu1991/GPT-SoVITS-VC。一切都保持不变，除了python vc_webui.py 编辑： 中文演示： [https://github.com/huangxu1991/GPT-SoVITS-VC中文演示：https://github.com/huangxu1991/GPT-SoVITS-VC/blob/main/docs/en/README.md](https://github.com/huangxu1991/GPT-SoVITS-VC%5B%E4%B8%AD%E6%96%87%5D(https://github.com/huangxu1991/GPT-SoVITS-VC)%E6%BC%94%E7%A4%BA%EF%BC%9Ahttps://github.com/huangxu1991/GPT-SoVITS-VC/blob/main/docs/en/README.md)

我的无法运行呢，我用的是集成包GPT-SoVITS-beta0217下载好您的集成包之后点击为什么python vc_webui.py弹出命令窗口然后立马关闭

https://github.com/RVC-Boss/GPT-SoVITS/assets/152373957/485c88cf-9ea6-462e-9c20-2ac98094948a 这是运行情况

重新拉取代码，再双击 go-webui.bat 试试呢

[Crinstttina]

大佬，请问能否在readme.md里面描述一下如何在colab里面使用您的变声功能呢？

[huangxu1991]

可以参考 https://github.com/huangxu1991/GPT-SoVITS-VC 移植到 colab 试试，也欢迎 commit！

[recluse90s]

Thanks for your VC project!

It's always on processing status when I try it by my own model, and there is no any log output in terminal, and it works when I use the python webui.py. I don't know why it not work. Thanks in advance for your help! @huangxu1991

Exec Env: MacBook Pro, M1 Pro, 16 GB, macOS 14.4.1, Python 3.9.18