mint quartz
tl;dr: V2 voice is great but behavior during singing often does not match the vibe/ atmosphere of the song. So would it be possible to add behavior presets for each song so that Neuro can tell if a song is "Dark, melancholic, funky, etc..." so that she can modify her behavior in a way that fits with it?
So, I love the new V2 singing voice! It's definitely amazing and is clearly an improvement over V1. Essentially, it sounds way more expressive and fluent (if that makes any sense). A new addition that also really like is that Neuro can now vibe to the song that she's singing. However, this is where the title of this post comes in. Sometimes (at least for me), it feels as though the way that she is vibing to the song doesn't match. An example of that would be during the first v2 karaoke stream where she was singing "Perfect Machine". Her interpretation sounded really good, however, most of the time, I ended up I g to look away from the stream (At other things) and just letting the voice sit in the background because her model would headbang in too much of an energetic way to a song that clearly was fairly dark in its nature. Essentially, the vibe of the singing and visual show did not match. Therefore here is my suggestion: (I'd like to preface that I'm no programmer or anything but) would it be possible for Neuro to have different preset behavioral toggles depending on the songs that she is singing? What I mean is to assign each song its own rules of behavior so that on one, she can act funkier on one, and on another, have a more aggressive demeanor.
For the programmers out there, do you guys see anyways in which this concept could realistically be implemented?