8 – Micromodulation149When Chowning tried to imitate the the singing voice by means of
electronic synthesis, he found that the impression of the lively quality of
the human voice is only achieved if periodic and random vibrato is added
to the sound spectrum. (Chowning, 1980). The importance of micromodu-
lation is confirmed by other researchers. Risset and Wessel (1982) conclude
that systematic as well as unpredictable variations of the timbral spectrum
are essential cues for naturalness. Carterette (1989) states that "jitter and
nonlinearity may be at the heart of musical perception. If a sound is too
pure it has no musical role."
Besides providing naturalness, micromodulation is a decisive factor for
the perceptual fusion of a soundstream and the segregation of simul-
taneous soundstreams. Mc Adams (1982) has investigated the effects of
periodic frequency modulation (vibrato), random frequency modulation
(shimmer) and very slow frequency modulation (portamento) as found in
inflectional changes in the voice or expressive pitch changes in musical
instruments. He concludes that coordinated modulation of spectral
components in the form of vibrato, shimmer or portamento is a strong cue
for the fusion of complex tones. Risset (1986) confirms that micromodu-
lations contribute to fusion if they are synchronous, to separation if they
are not.
The perceptual fusion of a soundstream by the coordinated micromodula-
tion of its spectral components is essential for the perception of the quality
of a particular instrument and the distinction of this instrument from other
instruments. The specific micromodulation of timbre in an instrument is
the basis for recognition of that instrument, and the particular micromo-
dulation of a human voice is the basis for the recognition of that voice as
the voice of one particular individual person.
Fig. 8.1. Nine listening dimensions