How to detect a voice sung over music

Question

Hello,

I'm looking to make a game where the player has to sing over a music that comes out of the device (no headset or headphones). The idea is that Unity detects only the voice (the attack and the note length, not especially the pitch) and that according to this detection it is able to know if the player sings in rhythm or not

I haven't really investigated yet, but I imagine that with a phase opposition, or a filter we could detect only the singing....

I would like to know if someone has already done this kind of game with Wwise or if you have any ideasI could search, article, video...

Thanks

Théophile D.