>It's not possiblme so far to extract pitch and timbre
information of
>seperate instruments out of a polyphonic audio source.
It's doable for a
>single monophonic voice (there is some software that can
do this pretty
>well). Beat detection isn't all that easy either.
Current beat detection
>algorythms seem to do well on dance music and related
styles, but often
>fail on anything else.
This sounds like a job for a neural network to me - they're
generally quite
good at pattern recognition problems of this type. I have no
idea if
anyone's tried it (I'm sure they have) and if so whether
they had any
success, but it's certainly the first thing I'd try.
Danny
_______________________________________________
dirGames-L mailing list - dirGames-L nuttybar.drama.uga.edu
http://nuttybar.drama.uga.edu/mailman/listinfo/dirgames
-l
|