Music Listening, Music Therapy, Phenomenology and Neuroscience

5 – Microtemporal listening dimensions: Timbre, Harmony and Pitch height

71

express a sinuous mountain silhouette by using portions of circles. In

fact, it is thousands of times more complicated. The intelligent ear is

infinitely demanding, and its voracity for information is far from

having been satisfied. (Xenakis, 1971)

The sensation of timbre is a joyful challenge to the intelligent ear, and the

description of timbre is a challenge to researchers in psychoacoustics.

Jean-Claude Risset and David Wessel (1982) have proposed a solution

to the problem of describing the transient part of timbre in their method of

analysis and synthesis. In the mid-sixties, Risset was working on computer

synthesis of brass-like tones. A first attempt was to synthesize tones with

fixed spectra of partials derived from analyses of trumpet tones. These

synthesized tones proved unconvincing when compared to natural

trumpet tones.

The next step was to record musical fragments played by a professional

trumpet player and analyze the trumpet sound in the form of spectro-

grams, visualizing the partials of the sound spectrum and the relative

predominance of certain frequency areas. The spectrograms showed that,

for a given intensity, the trumpet sound has a formant structure, that is, the

partials lying within a certain frequency range are enhanced as a result of

the characteristic resonance of the instrument. A peak in the frequency

spectrum was found between 1000 and 1500 Hz.

As a third step of the exploration, selected trumpet tones were

converted to digital form and submitted to a type of computer analysis

that yields a display of each partial as a curve showing the growth and

decay in time of that partial. On the basis of such an analysis, artificial

trumpet tones were then produced by a sound-synthesis computer

program. The resulting synthetic tones proved undistinguishable from the

original trumpet tones, so it was concluded that the third step of the

analysis and synthesis procedure had captured the aurally important

features of sound. A diagram of these features is shown in Fig. 5.1.

ng.