Information Technology Reference
In-Depth Information
experiences. In particular, he introduced the notion of
sound object
. The proposition
of P. Schaeffer naturally conducts the acoustician from the representational-com-
putational paradigm to the enactive paradigm, since P. Schaeffer in line with the
phenomenological viewpoint stresses the fact that sound perception is not only
related to a correct representation of the acoustic signal. This is also coherent with
later works of Varela and the conception of perception as an enactive process,
where the sound and the listener constitute a unique imbricated system. The per-
ception of the sound is modi
ed by the intentionality of the subject directed towards
the sound, which can induce an everyday listening, which is a source-oriented kind
of listening, or musical (or acousmatic) listening, which involves the perception of
the quality of the sound (Gaver
1993a
,
b
). Thus, sound synthesis should not be
limited to the simulation of the physical behaviour of the sound source. In other
words, it is the sound object given in the process of perception that determines the
signal to be studied, meaning that perception has to be taken into account during the
signal reconstruction process.
In the work of P. Schaeffer, morphology and typology have been introduced as
analysis and creation tools for composers as an attempt to construct a music
notation that includes electroacoustic music and therefore any sound. This typo-
logical classi
cation is based on a characterization of spectral (mass) and dynamical
(facture) pro
les with respect to their complexity and consists of 28 categories.
There are nine central categories of
sounds for which the variations are
neither too rapid and random nor too slow or non existent. Those nine categories
include three facture pro
“
balanced
”
les (sustained, impulsive or iterative) and three mass
pro
les (tonic, complex and varying). On both sides of the balanced objects in the
table, there are 19 additional categories for which mass and facture pro
les are very
simple/repetitive or vary a lot. This classi
cation reveals perceptually relevant
sound morphologies and constitute a foundation for studies on intuitive sound
synthesis.
Based on these previous theoretical frameworks from cognitive neuroscience, we
suggest that the control of sound synthesis can be discussed in the framework of the
representational-computational and the enactive points of view. In the approach
inspired by the representational-computational framework, we consider that the user
controls physical or signal parameters of the sound with the idea that the more
actual (with respect to the physical reality) the parameter control, the better the
perception. The physical or signal properties of sounds are considered as the ref-
erence for the sound control. In the approach inspired by the enactive framework,
we consider that the user is involved in an interactive process where he/she controls
the sound guided by the perceptual effect of his/her action. The idea is that the more
recurrent (and intuitive) the sensorimotor manipulation, the better the perception.
The sound control enables the perception to become a perceptually guided action.
This is an enactive process because the sound in
uences the control effectuated by
the subject and the control action modi
es the sound perception. The sound as
perceived by the subject is thus the reference for the sound control. Such enactive
framework formed a theoretical basis for a recent research community centred on
the conception of new human
-
computer interfaces (Enactive Network) and in a
