[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Shaeffer's Traité des objets musicaux -- was noise classification
I agree with your comments. I wish there were more who were thinking
along this line.
At 06:34 AM 04/17/2004, you wrote:
I have found this to be (like Helmholtz and Schaeffer) a melange of
cognition and perception that fails when there is an attempt to apply
it 'rigorously' in a "meaningful" (non-trivial) way. (The ability to
apply the results of the analysis as a creative tool.) But that may
be a discussion for other places.
Part of the recent history of this attempt to classify may be the
search for 'the' "unified field theory" of (sonic) perception and
cognition, including the drive to develop some kind(s) of analytical
tools that will allow the 'final classification' of all sounds /
You might be interested in my attempt at an operational analysis
of auditory perception. It is based on fundamentals of why and how hearing
has evolved as a sensory mode for aiding survival within the animal
kingdom. It brings in elements of meaning as well as the requirements and
physical limitations that are imposed by the acquisition of acoustic
information that is needed for survival.
My objective in doing the analysis was to develop a basic structure for
computational auditory scene analysis. I reckoned that achieving a CASA
model could best be done by understanding the essential functions, physical
requirements, and constraints of natural auditory systems. (I would welcome
a reference to any such pre-existing analysis.) By knowing the limits of
what is required it becomes possible to design and implement an ASA system
according to attainable parameters.
The structure that has evolved from my analysis has similarity with
mammalian auditory pathways, yet without any overt attempt to replicate
them. If this analysis is correct, it could explain current difficulties in
auditory modeling. In any case, it does intend to explain why a siren at
3:00 AM is noise but not noise in an emergency, and why 12-tone atonal
music has an irritating percept. (At least, it does to my ears.)
Even without touching the voice, I find that the classification of
(This is (?) a psychometric definition of noise rather than an
acoustical one (?).) Therefore the classification system needs to
acknowledge the two parts (perceptual and cognitive -- the sensation
/ what the sensation is interpreted to mean).
These factors are considered in my analysis.
There are many other layers to this thread, to be followed up in other places.