Yep. Using AI at all here feels like overkill. You could get interesting results just choosing random chords* in the key signature.

*speaking from experience. I built a web app at tapcompose.com (desktop only for now) that uses pure randomness for both chords and melody. Initially I used randomness for testing only and planned to substitute an AI model later — but I was delightfully satisfied with the results and kept it in

I cheat a bit and use Scaler 2's library as a starting point. It has progressions for all kinds of moods and genres, and that's usually what I'm aiming for.

Ooh TIL about plug-ins like Scaler 2. I need to get more immersed in the DAW world.

It'd be cool to train text embeddings and musical chord embeddings (by analyzing song lyrics perhaps) and then build something to type in moods like "happy jaunty progression" and get a matching chord progression with a similar embedding (assuming you give it a large library of progressions). No generative AI would be needed :)

But it produces the worst, results. Just examine some music that you like an you will find it is far more complex than that. Often a single line of notes is supported by the chords rather than being shaped by them, and the chords can vary wildly in duration.

A chord is two or more different notes being played close together in time. That's all.

It's a small tool for generating variations on a chord progressions, not writing songs or coming up with melodic ideas. No need to make everything complicated.