Subject: Re: [linux-audio-user] EVSynth - genetic algorithms and sound synthesis
From: Dave Griffiths (dave_AT_pawfal.org)
Date: Wed Feb 20 2002 - 18:45:54 EET
> One I considered a few years back is similarity: have an input sound at
> different pitches and the pitch and the metric is the similarity of the
> output sound and input sound.
>
> That way you could have a kind of learning synth. I'm not sure how well it
> would work, but it could be interesting.
It's possible to get a neural network to learn a pattern, and then
extract the parameters beyond the range of the original data and use it
as a prediction. That could be done with a piece of music too - write a
small section of a song and get the net to expand it. Maybe with
spectacular results... maybe it would be awful ;)
The neural net plugin is not really capable of that kind of thing
though, it "learns" through having it's structure and weights evolved in
the same way as the l system (by human decision).
I've had some fun with it though, mainly semi random drum loops, that
shift around for minutes on end.
Dave
: www.pawfal.org :
This archive was generated by hypermail 2b28 : Wed Feb 20 2002 - 18:35:11 EET