search

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

Easily train a good VC model with voice data <= 10 mins!
MIT License
24.61k stars 3.61k forks source link

Bad quality at the end after converting #1630

Closed ringi1970 closed 6 months ago

ringi1970 commented 11 months ago

I tested RVC-fork with a good tutorial on Youtube from "AI Tools Search". I have a RODE NT1A microphone and spoke (recorded with Presonus Studio One) one short sentence and took a little bit EQ on it. Then downloaded a free "John Lennon" voice and converted it. The output wave file sounded like it is a 64bit mp3 file. Very bad quality, no depth Then i trained 5 hours an own Lennon voice (seperated his voice from "A Working Class Hero" with 5000 epochs), but the result sounded also like it was a bad mp3 file. What did I wrong?

Thanks for help

THE-SINDOL commented 10 months ago

You have few things to do on your mp3 file. https://rentry.co/TrainingVoiceModels https://rentry.org/RVC_making-models

github-actions[bot] commented 6 months ago

This issue was closed because it has been inactive for 15 days since being marked as stale.