Information Technology Reference
In-Depth Information
ing (DSP) algorithms, or a set of parameters for
existing sound objects, rather than actual sound
files. The product is “potential sound”, rather
than particular audio data as it would appear at
the digital analog convertor (DAC). The complete
package of a sound object is DSP, control, and
encapsulation code compatible with a set of pa-
rameters to be supplied in the future. Objects may
be instantiated and animated at some later time,
in any contrived circumstance. We say this media
has deferred form , exhibiting desired behaviour
according to supplied runtime parameters.
sation of models at an abstract level is possible,
perhaps to combine fluid models like bubbling
mud with fire to create lava flows.
If procedural audio implies the sound is based
on process, then behavioural audio implies that the
internal model and supplied parameters reflect the
behaviour of the target object in some way. What
is behaviour? Behaviour relates environmental
stimulus to what is observed in time. It is a strange
concept because it must both change and, to some
degree, remain fixed in time. A purely signal in-
terpretation might define behaviour as repeatable
changes in one or more output features in light of
one or more supplied, dependent variables with
little restraint on the whole cavalcade of qualifiers,
discrete, continuous, linear, non-linear, causal, or
non-causal that might apply. A useful behaviour
might be supposed to be time invariant on a me-
dium to long scale (in the order of seconds). In
other words, behaviour that changes too rapidly
ceases to be observed as behaviour.
When talking about behaviour we might
incorporate statefulness into sound objects, for
example; whether a container is full or empty
will cause it to respond differently to a collision.
Within a larger context behaviour implies a nar-
rative. A drink can or tumbleweed in the street is
destined to roll around as it is blown in the wind.
Of course it can do many things, but rolling around
is its “script”, its purpose in life. One way of un-
derstanding the narrative is a context that places
particular emphasis on certain features. In real
life the function of glass windows is to keep out
the wind while allowing light in. In a first person
shooter their primary concern (understood tragic
destiny) is to be shot at and broken. As we will
examine later, this change of narrative focus (and
thus reality) is both a dynamic force in interac-
tive sound, and the source of an error in some
programming approaches that assume the need
for a literal, uncoloured interpretation of reality.
Procedural Audio as a
Design Philosophy
It's worth adding that the above goal is not the
exclusive end. Merely adopting a procedural way
of thinking about sound can inform and enhance
traditional design approaches. For this reason I
teach a sound design syllabus based upon this
premise. Students begin the first semester by
considering the physical processes inherent in
all sound sources They then progress to devising
models and appropriate synthetic methods in an
implementation language. This deeper under-
standing enables choices and creativity through
structural metaphor and simile rather than only
empirical (surface) features.
For example, we study the design of bells
with reference to design traditions on shape and
material properties. Combining metallurgy and
geometry (which leads to modal interpretation)
with the study of basilar physiology and sensory
psychoacoustics of Barkhausen/Zwicker (1961)
and Plomp and Mimpen (1968), we arrive at a firm
understanding of why some bells are sonorous, dis-
sonant, hollow, or foreboding. In another exercise
we decompose the sound of fire to arrive at the
components corresponding to physical processes
of combustion. This results in models of fire with
surface parameters like crackles, hiss, and roar,
which can be abstracted further into scalable fire
models for burning trees or liquid fires. Hybridi-
Search WWH ::




Custom Search