I believe this is only effectively possible, not fully possible. Inherently, music and the voice will share some of the same frequency samples, since they are d

I believe this is only effectively possible, not fully possible. Inherently, music and the voice will share some of the same frequency samples, since they are discrete. I’m sure you might be able to get a solution that works to the human ear, but I don’t know that it’s possible to perfectly strip out one or the other.