additive

[Chris Stecker]

The difficulty with the analysis-synthesis approach you describe
comes from the large number of oscillators necessary to produce
a convincing reproduction of a natural acoustic sound.  Anywhere
from 50 to 300 or more may be necessary, depending on what other
processes you use to produce transients, noises, etc.  This tends
to be more oscillators than found in modular analog synths, so
the obvious solution is to go digital, emulating these oscillators
in software.

This approach has been studied extensively at Berkeley's CNMAT
(http://cnmat.berkeley.edu), where they have used SGI machines
to do real-time (controllable!) digital additive synthesis.
With the advent of MSP for PowerPC Macintoshes, similar techniques
can be achieved with more modest computational setups.
(see http://www.cycling74.com about MSP)

The main difficulty in doing this is not the synthesis itself
(if you know the freq, amp, and phase data, it can easily be
programmed), but the analysis.  You can take an FFT of a signal,
which will show you its spectral makeup, and from this you can
identify the main harmonic peaks (for harmonic sounds).  But it's
the quiet stuff that really makes a difference in the realism of
the simulation, and these quieter components can be difficult to
pick out.  In real instruments, many of these may be inharmonic,
making the analysis procedure even more difficult.

Of course you can take the FFT representation, perform an IFFT
and resynthesize your original signal.  With MSP, you do both
the FFT and the IFFT in real-time, modifying the data in between.
You can even (as mentioned above) design procedures that try to
match an additive-resynthesized version to the real signal.  If
you have a Power Mac, and about $600 for the software (Max + MSP),
you're ready to try some experiments.  I can not recommend this
program(ming language) highly enough for prototyping ideas like
this.

-Chris Stecker



>hi guy's.
>theoretcly any acostic sound can be made in a modular
>ether by addtive or FM but this is hard to be made.
>so my qustion is can i play a acostic instrument like a
>pino or violin or other in to a mic and then in to a scope
>and see its waveshape and envelope and then calculaite
>this data and creat the same sound on a modular ?
>all so is ther any scope or other unit that can take a waveshape
>and do the calculate and give a data of how many sine osc in
>what freq, amplitude and phase do i need to creat the
>same waveshape.
>or even a unit that is cv conected to sin vco's and automaticly
>creat the waveshape.
>
>thanx
>Gur Milstein


___________________________________________________________________________

Chris Stecker

cstecker at cogsci.berkeley.edu

Graduate Student, Psychoacoustics
3210 Tolman Hall, #1650
University of California, Berkeley
Berkeley CA 94720-1650

Auditory Lab, B-50 Tolman Hall, (510)642-5352   http://ear.berkeley.edu

!!Ask me about Space Mesa, Ovenguard Music, Receptacle Culture, and CELL!!