Yes, it's scary-good sometimes. laugh

But... you can make it even better with some intervention. For example, the riff at 0:26 doesn't really fit in with what happened before it. At at 2:00, there's a similar jump. At 2:16, it sounds like two different ideas are spliced together.

It's sort of like generating poetry with a Markov chain - at the micro level, it works, but there's little connectivity once you move from sentence to sentence.

But if you took that into a DAW and edited it so phrases repeated, and enforced some structure on it?

Add a little human intelligence to that AI, and you've got some even sweeter music. wink


-- David Cuny
My virtual singer development blog

Vocal control, you say. Never heard of it. Is that some kind of ProTools thing?