Say you are playing quarter-notes for one measure at 92 bpm, next at 88 bpm, next at 95 bpm and next at 95 bmp.
Ok, suppose you have bad rhythm. Most people can still detect what rhythm you are trying to keep. How do they do it? It's complex and requires intelligence. Rhythm induction is not a simple problem. That's why it falls under AI.
In any case, if the machine knows the tune that you're trying to play (e.g. has a representation of it in a MIDI file), it can do a much better job. Just like in a human accompanist. Even better would be it if it had a representation of expressive performance (i.e. intentional tempo changes).
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
krja
(no subject)
Date: 2005-11-11 10:21 am (UTC)Ok, suppose you have bad rhythm. Most people can still detect what rhythm you are trying to keep. How do they do it? It's complex and requires intelligence. Rhythm induction is not a simple problem. That's why it falls under AI.
In any case, if the machine knows the tune that you're trying to play (e.g. has a representation of it in a MIDI file), it can do a much better job. Just like in a human accompanist. Even better would be it if it had a representation of expressive performance (i.e. intentional tempo changes).