If you care about music: at least 100 years, almost certainly much longer. I don't think any of us in this comment section will live to see an AI that truly understands human music.

If you care about money and don't mind making the world a more terrible place: humiliate a few dozen human classical pianists by making them record hundreds of hours of motion capture, invest in engineering a good robo-arm, and I would guess in 5 years you'd have something passable.

I am skeptical to a robot's ability to interpret a piece of music from say Gershwin, Bernstein, Mozart or Beethoven. The beauty of human musicians is that each high level musician my have his own approach to a given piece of music. It is not just about the scores written but to interpret what the composer might have thought or felt. A robot playing these pieces might send the a.m. composers spinning in their graves.

Some new Steinway pianos have a 'high resolution player piano' feature where the performances of concert pianists can be played back on your Steinway. Not really a robot soloist, but I thought it was interesting. Not interesting enough that if I suddenly came into Steinway money I'd get one with that feature, though.

record players are perfect

robot lips are pretty poor. They exist, but not at the point of making brass/wind instruments practical.

most other things can be operated adequately by robots.

But then, you're better off with either synthesisers or direct genAI synthesis.

However I do wonder what the point of offloading the distillation of human emotion to robots is. The whole point of music is that it contains what someone _feels_.