[GH-ISSUE #1351] Training VCTK from scratch #876

New issue

Open

opened 2026-10-20 10:03:53 +00:00 by bsakaguchi · 3 comments

bsakaguchi commented

2026-10-20 10:03:53 +00:00

Owner

Originally created by @Rolun on GitHub (Oct 2, 2023).
Original GitHub issue: https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/1351

Hello,

I am training the model on VCTK from scratch.
Is there anything I should keep in mind?
For example, should I use both mic1 and mic2, and all 110 speakers?

Many thanks in advance!

Originally created by @Rolun on GitHub (Oct 2, 2023). Original GitHub issue: https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/1351 Hello, I am training the model on VCTK from scratch. Is there anything I should keep in mind? For example, should I use both mic1 and mic2, and all 110 speakers? Many thanks in advance!

bsakaguchi added the

following up

label

2026-10-20 10:03:53 +00:00

bsakaguchi commented

2026-10-20 10:03:54 +00:00

Author

Owner

@RVC-Boss commented on GitHub (Oct 9, 2023):

I used the 108 speakers version.

@RVC-Boss commented on GitHub (Oct 9, 2023): I used the 108 speakers version.

bsakaguchi commented

2026-10-20 10:03:54 +00:00

Author

Owner

@Rolun commented on GitHub (Oct 9, 2023):

@RVC-Boss I've only seen the 110 speaker version so far, do you have a link to the 108 version?

@Rolun commented on GitHub (Oct 9, 2023): @RVC-Boss I've only seen the 110 speaker version so far, do you have a link to the 108 version?

bsakaguchi commented

2026-10-20 10:03:54 +00:00

Author

Owner

@KushATrivedi commented on GitHub (Feb 21, 2025):

Hi @Rolun @RVC-Boss !
I want to train the model from scratch on my own dataset instead of VCTK. Could you guide me on how to proceed with this? Also, is there's any documentation available on this process?

@KushATrivedi commented on GitHub (Feb 21, 2025): Hi @Rolun @RVC-Boss ! I want to train the model from scratch on my own dataset instead of VCTK. Could you guide me on how to proceed with this? Also, is there's any documentation available on this process?