[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: noise classification problem



Back in 77, Pierre Schaeffer wrote very interesting things about sound (including "noise") classification (what he called "typo-morphology") in a book called "Traité des Objets Musicaux", 'Treatise of Musical Objects'.
It is written in French.
There might be some translations of this work in English,
for e.g. http://www.sun.rhbnc.ac.uk/Music/Archive/Disserts/palombin.html
but I am not aware of any official edited translation.
Note, that it was thought as a tool for musical composition (mostly electroacoutic music) which might be slightly out of your scope (?)

regards,
vincent

At 22:27 15/04/2004, Valeriy Shafiro wrote:
Hi Alberto,

I don't believe that there are any "official" categories for classifying
real world sounds.  In my opinion, Gaver's taxonomy of environmental
sounds, while clearly not perfect, is still the best that we have for
classifying sounds in general.  At least it is a great starting point.  As
you wrote in your email the problem of classification is very complex, and
this is one reason why you have not been able to find much information
about it.  Real world sounds are produced by a great variety of different
sound sources which cannot be unambiguously classified either.  People have
tried to find some kind of an underlying perceptual structure of
environmental sounds (e.g., Ballas, 1993; Marcell et al., Gygi, 2001), but
that has not revealed any clearcut categories.  Which is not to say that
there is no category structure, but rather that the categories do not
reveal themselves very easily and unambiguously with the analysis methods
we are using.  My preferred analogy for the perceptual organization of real
world sounds would be that of the lexicon where individual items can be
classified based on acoustics/phonology, and also based on the ecological
significance/semantics/meaning.  Of course, this analogy is not perfect,
and I offer it just as one way to think about the problem.  For one, for
most environmental sounds the relationship between their semantics and
acoustics is not as arbitrary as it is for words.

If I understood you correctly, and your goal is synthesizing musically
useful noises (possibly based on some real world sounds) then rather than
trying to come up with a general all-encompassing classification of real
world sounds you may have more success figuring out specific types of
noises/sounds that maybe interesting for your application.  Or, you can try
to find a way to represent different types of sounds in a smaller subset.

Best regards,

Valeriy
-------------------------------------------------------------
Valeriy Shafiro
Communication Disorders and Sciences
Rush University Medical Center
Chicago, IL

office (312) 942 - 3298
lab    (312) 942 - 3316
email: valeriy_shafiro@rush.edu

Refs:

Ballas, J.A. (1993). Common factors in the identification of an assortment
of brief
everyday sounds. Journal of Experimental Psychology: Human Perception and
Performance, 19 (2), 250-267.

Gygi, B. (2001). Factors in the Identification of Environmental Sounds,
Unpublished
doctoral dissertation, Indiana University, Bloomington, Indiana. Retrieved
02/20/02
from http://www.indiana.edu/~k300bg/dissall.pdf

 Marcell, M.M., Borella, D., Greene, M., Kerr, E. & Rogers, S. (2000).
 Confrontation
naming of environmental sounds. Journal of Clinical and Experimental
Neuropsychology, 22(6), 830-864.