杨永全
杨永全
我用了一分钟左右的声音微调的效果也不如直接用底模推理的声音好。不知道是什么原因。
the piper-phonemize does not support python 3.11.
补充一下,修改 onnx_export.py的第七行,将 `cnhubert_base_path = "pretrained_models/chinese-hubert-base"` 改为 `cnhubert_base_path = "GPT_SoVITS/pretrained_models/chinese-hubert-base"`, 文件末尾的大约329到331行,改为你自己的模型名字。 然后后回到上层目录运行: `python GPT_Sovits/onnx_export.py`
可以使用下面这个项目调用 onnx : https://github.com/NaruseMioShirakana/MoeVoiceStudio
请教下有没有12306查询车站信息和中转信息的接口呢?
谢谢。找不到不要验证码的吧?
Thanks for @lpscr. follow your work , I can train on 4060Ti.
Maybe you can use the follow command to install the piper-phonemizer: ```bash pip install piper_phonemize -f https://k2-fsa.github.io/icefall/piper_phonemize.html ```
Very concerned about progress