Abstract
One-third-octbandfilter spectra of nine different vowels, each obtained
as theaverage spectrum of vowels sung by professional male and femalesingers
in nine modes of singing, were obtained at sixfundamental frequencies
(F0), ranging from 98–880 Hz. For F0=98 Hzboth an empirical and a
theoretical comparison were made betweenthe representation of vowels
in the formant space and therepresentation in the spectrum space.
A first-order correspondence between bothapproaches was demonstrated.
The vowel configurations in the spectrum spaceshowed that, apart
from differences in the grand-average spectrum, (1)for F0<220 Hz,
vowel configurations were similar for males andfemales; (2) for F0=392
Hz, the variability in dimensions relatedto F1 and F2 was smaller
than for lower F0values, and variability related to F1 in male falsetto
registerwas greater than in modal register (vowel spectra were similarin
male falsetto register and female singing); (3) for F0=659Hz, the
vowel configuration had shrunk to clusters of frontand back vowels;
and (4) for F0=880 Hz, vowel differenceswere only marginal. The relation
between the average sound levelof the so-called singer's formant
and F0 appeared to bevowel dependent. Up to F0=392 Hz, the singer's
formant wason the average equally prominent for male and female singers,but
for higher F0 its level, relative to overall sound-pressurelevel,
dropped for females. Differences between average spectra of sungand
spoken vowels are discussed.
Users
Please
log in to take part in the discussion (add own reviews or comments).