computer models of the flute

From: joel taylor (joel.g.taylor@comcast.net)
Date: Sun Sep 21 2003 - 15:13:30 PDT


  Re Phil's comment about samplers. That's not the way to go for a
good computerized flute sound. The way to go is to create a
physical model of the flute, and then have that model (a set of
equations) compute the sound. This is a common technique now, and
there ARE good physical models for the transverse flute available. I
have one running on my laptop.
One of my favorite computer instrument models is a combination of the
equations for the flute with those for the electric guitar, the
creator calls it Blotar. Heh.
You can create a combination of a flute 12 ft long with a tiny
ukelele line electric guitar, if you wish.
Anyway, modeling the shakuhachi doable...by programming an adequate
physical model for the players embouchure and the resulting wind
stream as well as a model for the acoustic impedance of the
instrument proper, you might be able to really do a good computerized
shakuhachi, at least theoretically.

The problem is the same as with most electronic instruments...the
huge stream of numbers you will have to feed the model to simulate
what we do with our breath and embouchure, and fingers, is the part
that usually makes such models less interesting, musically, than the
instruments they model...however, theoretically, it's just a matter
of getting the equations right.
And from looking at that australian paper on the impedance funtion of
the shakuhachi in combination with the players face, i'd say the
right equations are out there.

Samplers are just recorders, they can't model anything.

One point. Even if someone puts together such a model, that does not
mean that they will be able to figure out how to feed it well enough
to make it sound like a flute in the hands of even a good player,
much less a great one.
On the other hand, i'd like to hear what a 20 ft shakuhachi would
sound like, just playing Ro and Ro Dai Meri.

I wouldn't be surprised to find out this has already been done, actually.
Digital Signal processing has come a long way in the last ten years.

cheers, and
peace to all,

joel

-- 

Joel Taylor

New Email Address: joel.g.taylor@comcast.net _____________________________________________

List subscription information is at: http://communication.ucsd.edu/shaku/listsub.html



This archive was generated by hypermail 2b29 : Tue Jan 06 2004 - 14:09:35 PST