AI can generate the sound. The artist decides why it exists. — Studio 2.0
Studio 2.0 accidentally taught me something about production.
While working on a song I was previously trying in legacy and crossed over to the new version, I layered three versions of my vocal: lead centered, with two quieter doubles panned left and right at about -12 dB and given reverb.
Somewhere in the process, one of the doubles got slightly time-shifted/warped. Basically, I got overwhelmed(still learning.)
But when I listened back, the offset stereo vocals + reverb created this strange temporal quality—like different versions of the voice were arriving from slightly different moments rather than simply occupying different places in the stereo field.
And it fit the song.
That’s when something clicked:
AI can generate the sound. The artist decides why that sound exists.
It gave me enough control, (and the new plugins for reverb I’ve been dying for)— to hear an accident, understand what it was doing emotionally, and decide that the “mistake” belonged in the finished record.
That’s what excites me about 2.0. The generation doesn’t have to be the finished product anymore. It can be material—something we separate, move, layer, distort, mix, and give intention to.
Out of reverence for my song, I ultimately left some of its imperfections intact. They felt more appropriate than polishing the life out of it.
I’ll leave the track below for anyone curious to hear what I’m describing.
[Dark Cinematic Funeral Anthem] “Ash Testament” by Decoherence