[GH-ISSUE #361] Getting weird accents #175

New issue

Open

opened 2026-10-20 09:56:20 +00:00 by bsakaguchi · 2 comments

bsakaguchi commented

2026-10-20 09:56:20 +00:00

Owner

Originally created by @kainbock1 on GitHub (May 27, 2023).
Original GitHub issue: https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/361

The voice of the audio files processed with my model sounds like me, but with strange accents. Can only certain languages be spoken/sung cleanly? Or did something go wrong during training or something?

Originally created by @kainbock1 on GitHub (May 27, 2023). Original GitHub issue: https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/361 The voice of the audio files processed with my model sounds like me, but with strange accents. Can only certain languages be spoken/sung cleanly? Or did something go wrong during training or something?

bsakaguchi added the

help wanted

documentation

labels

2026-10-20 09:56:20 +00:00

bsakaguchi commented

2026-10-20 09:56:20 +00:00

Author

Owner

@fumiama commented on GitHub (Jun 1, 2023):

Maybe because the pre-model was trained by the dataset that isn't in your language. Increasing your epoch count may help to fix that.

@fumiama commented on GitHub (Jun 1, 2023): Maybe because the pre-model was trained by the dataset that isn't in your language. Increasing your epoch count may help to fix that.

bsakaguchi commented

2026-10-20 09:56:21 +00:00

Author

Owner

@kainbock1 commented on GitHub (Jun 1, 2023):

Maybe because the pre-model was trained by the dataset that isn't in your language. Increasing your epoch count may help to fix that.

2023-05-25 14_21_57-Gradio – Mozilla Firefox

These are the Setting I tried. What changes would you suggest? What would be a good epoch count? And are there other pre-models I could use? :)

And thank you for replying!

@kainbock1 commented on GitHub (Jun 1, 2023): > Maybe because the pre-model was trained by the dataset that isn't in your language. Increasing your epoch count may help to fix that. <img width="1146" alt="2023-05-25 14_21_57-Gradio – Mozilla Firefox" src="https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/assets/115891670/057b38de-2a11-4983-8d5b-8116486f8626"> These are the Setting I tried. What changes would you suggest? What would be a good epoch count? And are there other pre-models I could use? :) And thank you for replying!