Jump to main navigation


Whole-Play update

080608

I seem to be a bit stuck on the harmony analysis task, it's been a long time since I started it, and there's still a lot to do. Well, I didn't expect teaching Chuck some harmony would be particularly easy... But still I need to keep moving on, so I've decided to close the module I was working on (harmonyDetector) and leave it as it is for now. Which is not too bad, although still far from my goal. I think I generally tend to be too much of a perfectionist, and could spend my whole life working on this single issue, so I'm 'forcing' myself to be imperfect. :)

At the moment Chuck can detect harmonies based on pentatonic, diatonic, whole-tone and octatonic (1t-1st or 1st-1t) scales. It also has a default 'generic' response for harmonies that don't fall into any of the previous scales (that's one of the parts that needs more work). The response in any case is totally free: any parameter can be influenced by the detected harmony, including rhythm, time signature, instrumentation, and of course the harmonic response itself.

Harmony transitions are also detected (with variable success, a lot of room for improvement here). But it's not bad at all, and Chuck's getting relatively quick to pick them up.

So before I go on to improve the harmony analyser, I've decided to start the second milestone. It's going to be based on the current harmony analyser, trying to make the most of the current harmonic possibilities. It's quite exciting, really. Compared to milestone 1, Chuck now responds in a more inteligent fashion, and I'm hoping the end result will be more satisfying than with milestone 1. I wanted to have it ready by Sunday 15th, but somehow I don't think that's going to happen... well, I can try at least.

You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

Comments to “Whole-Play update”

Comments are closed.

Additional content and navigation

Categories

Recent comments