[anigbrowl]

'Neural audio synthesis' really misses the point. This is just the same technique applied to emulating real instruments, which is already a massively well solved problem, and unlikely to result in any significant improvement.

Music synthesis, on the other hand, is an unsolved hard problem. For example, producing harmonic accompaniment to a hummed melody, or rendering a given chord sequence in different musical styles, or making a drummer that can pick up where someone's beatboxing left off..any of these have the potential to reach a huge audience.

Synthesizing real instruments basically says 'we know nothing about this space and are going to ignore all the works that's already been done in it.'

[jay-anderson]

> This is just the same technique applied to emulating real instruments, which is already a massively well solved problem

I disagree. Most instrument synthesis techniques that I've heard don't yet approach real physical instruments. I say let them try using these techniques to an old problem and see if they can get any improvements.

Automated musical expression and improvisation are also active areas of research. (One interesting example: https://www.youtube.com/watch?v=qy02lwvGv3U - though I wonder how much it is listening and responding vs how much is pre-programmed.)

[anigbrowl]

Inability to get a particular sound you want in a well-equipped audio production studio is not a problem in 2017.

[jay-anderson]

I have never heard emulation of a physical instrument that matched the real thing. For instance movie studios hire musicians to fill in full orchestras for that reason. If they could get the same result without the musician I'm sure they would.

[dunkmaster]

".. or rendering a given chord sequence in different musical styles"

Shameless plug for my RNN trained piano music generator which can do exactly that:

www.lawsonhe.com/music.html

It can also generate novel chord sequences and compose to those.