Yeah well, unfortunately this is based on magic (machine learning) so there isn't really any way to tune anything, pretty much by design. And I didn't even make the models myself, so that makes it even less possible for me to change anything about it.
In theory it would be possible for someone to try their own models and get them in there, but I haven't implemented a mechanism to supply your own models, and creating a model yourself is perhaps a bit too much work for most people.
It cleans up noise pretty well, but when I apply it to my mic track it causes approx. 1.6-1.7 second delay (it's permanent and doesn't increase or decrease over time) for computer audio track which doesn't even have this effect applied.
Hmm, I didn't get a notification, sorry for the late response. This is definitely not normal. Unfortunately these things can be quite hard to troubleshoot and this is a hobby project I don't have that much time to spend on, so I'm kind of doubtful I can help you resolve this. Sorry!
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
In theory it would be possible for someone to try their own models and get them in there, but I haven't implemented a mechanism to supply your own models, and creating a model yourself is perhaps a bit too much work for most people.