Re: robot com-posers


Subject: Re: robot com-posers
From: Ross Bencina (rossb@audiomulch.com)
Date: Thu Feb 17 2005 - 14:16:15 EST


Nick Wrote:
> MY idea is to focus on a certain type of music, as one would focus on one
> language at a time when teaching a computer how to speak. Build a big
> database
> of rock and roll and teach the computer to recognise key percievable
> elements
> of that type of music. I'm not saying its going to be easy.
>
> you would have to disect the songs to their elements. Teach it to
> segregate the
> drums from the guitar and so on, teach it to recognize modes progressions.
> rhythmic patterns, timbres... but on the other hand it would have to
> recognise
> the big picure: tempos, style, overall form,,,the list should be very
> long.
>
> Then the computer could design a piece using a huge matrix, each value
> being
> one of a million other values in its category but a problem emerges here.

This is already being done to a large degree. Music information retrieval
systems can extract these kind of descriptors from sound files. People have
used sequences of descriptors extracted from one piece of music to
reassemble fragments of other pieces of music into new peices with the same
descriptor sequence but different sound sources. It's called audio mosaicing
and doesn't require an "linguistic" understanding of the materials...

Ross.



This archive was generated by hypermail 2b27 : Sat Dec 22 2007 - 01:46:06 EST