Question about models
Hi! As I understand from the names in the models, F - Fullness, B - Bleedless, is that correct? If so, how can you explain the following files:
inst_gabox.ckpt
instrumental_gabox.ckpt
INSTV5.ckpt
Is V5 a continuation of the V4 Fullness model or something else?
Sorry for the possibly silly question, but I'm just curious to know :)
instv5 is focused on fullness.
instrumental and inst are the base "balanced" models for both fullness and bleedless
我也想知道 N 又代表什么呢，谢谢
这意味着这个版本的乐器噪音更多，但似乎更饱满、更好
And what about to train the new melband karaoke and multispeaker combination model? I used the old melband karaoke model but there's still back vocals.
Well, Gabox, is it your new karaoke model still experimental? Because that model can't remove well back vocals at all.
Well i'm not really interested on karaoke models so maybe that's the only version you'll have
Well, I tested your Karaoke model.
This audio that I separated, idk... It's still experimental, and it removes the crowd, sfx, and everything, but the back vocals...