Information Technology Reference
In-Depth Information
linked to the afferent concept in the same consubstantial way. For example, if I hear
a bell, I immediately know that it is a bell, and perhaps, but not always, I even
manage to imagine the size of the bell, depending on its spectral contents. Except
for
ed,
one can say that each sound can be considered as a non-linguistic sign whose origin
can be described using language, in a re
“
abstract sounds
”
, i.e., sounds for which the sources cannot be easily identi
ective thought. Previous studies have
shown that the processing of both linguistic and non-linguistic target sounds in
conceptual priming tests elicited similar relationships in the congruity processing
(cf. Sect.
4.5
). These results indicate that it should be possible to draw up a real
semiotic system of sounds, which is not the linguistic semiology, because pho-
nemes can be considered only as particular cases of sounds.
So far, the identi
cation of signal invariants has made it possible to propose an
intuitive control of environmental sounds from verbal labels or gestures. An
interesting challenge in future studies would be to propose an even more intuitive
control of sound synthesis processes that bypasses words and gestures and directly
uses a BCI that records electroencephalographic signals in a BCI/synthesizer loop.
This idea is not new and several attempts have already been made to pilot sounds
directly from the brain activity. In (V
ä
ljam
ä
e et al.
2013
), the authors made an
exhaustive review in the
cation in various applications (medical,
neurofeedback, music, etc.) and concluded that the type of mapping strategy
strongly depends on the applications. For instance, in the case of musical appli-
cations, the mapping is generally determined by artistic choices and does not
necessarily mirror a strict semiotic relation. The intuitive BCI-controlled synthe-
sizer that we aim at is intended for a generic context and should enable the iden-
ti
field of EEG soni
cation of brain activity linked to speci
c signal morphologies that re
ect the
attribution of sense to a sound.
This paper is organized as follows. In Sect.
4.2
, the methodology that leads to
intuitive sound synthesis is viewed in the light of representational-computational
and enactive perspectives. Then, in Sect.
4.3
, two sound synthesis approaches are
described and related to the previously presented perspectives. In Sect.
4.4
, different
control strategies emanating from the different synthesis approaches are described.
In Sect.
4.5
, some results from experiments supporting the existence of semiotics
for non-linguistic sounds are presented. Finally, in Sect.
4.6
, a prospective view on
a control strategy for synthesis processes based on a BCI is proposed.
4.2
Two Conceptions on the Way We Interact
with the Surrounding World
Sound synthesis that integrates perceptual effects from the morphology of their
signal in order to enable intuitive control to the end-user brings forward the fol-
lowing questions: How do I attribute a meaning to a perceived sound (related to the
semiotics)? What effect does this sound have on me? These questions induce a
change in our position with respect to the sound from a third-person position
