i have a dataset worth of 108 samples of DOOKIE era billie joe armstrong can someone do the training
i have made a dataset worth of 10 second clips from the album DOOKIE and i was wondering if someone could train it it would mean so much to me
Yes, I will be happy to train it. Send me the link
There you go and thank you ❤️
I will put to train right now. But I think it would be better if you remove any part that has back vocals or any case other than billie joe voice alone.
Yeah I'll have a look tomorrow since I'm not busy tomorrow
If I did put backing vox in
How is the progress on it so far?
I wasn't able to use the server I expected to use, so I will try on colab today.
There it is the first hour of trainning https://drive.google.com/file/d/1ST4thWhyKCH5mMc6Ht1OB85k1fwlw37U/view?usp=drive_link
Is it suppose to be just the g16600.pth?
you can use the same config.json of the others. i always use the same
Yea right
It looks already good. what do you think?
i haven't hada try of it yet lol my parents are being dicks about the wifi
I have got to have a try of it today and it sounds pretty darn great man thanks for helping
How do I use this with RVC? There is no index for this.