RVC-Project/Retrieval-based-Voice-Conversion-WebUI

How do I use pretrained model?

Open

#363 建立於 2023年5月28日

在 GitHub 查看
 (3 留言) (0 反應) (0 負責人)Python (18,427 star) (2,849 fork)batch import
documentationhelp wanted

描述

Sorry if it is a dumb question, but I can't seem to figure out which models are which. I followed the instructions for installing, which includes downloading from huggingface. However, the inference voice list is empty, even if I restarted the program. I am new to this so I might be missing something, any help is appreciated.

貢獻者指南

技術棧
python
領域
machine learningbackend
議題類型
documentation
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
2
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
under 1 hour
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
needs maintainer response
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
mostly clear
前置要求
無
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
70
研究方向
First, verify that the pretrained model files have been downloaded correctly from Hugging Face and placed in the expected directory. Then, check the configuration file (e.g., config.json) to ensure the model path is set properly. Look at the webui's model loading routine in the Python source code (likely in `modules/` or `infer.py`) to see why the voice list remains empty. Inspect any error logs for missing file exceptions or incorrect paths. If the issue persists, replicate the installation steps in a clean environment to confirm the documentation is accurate.