Return to search

A Spectral Analysis of Selected Vowels Sung by Bass and Baritone Student Singers

While a limited body of research pertaining to vocal sound quality exists, technological advances in sound analyses have facilitated a reexamination of vocal timbre. The sound quality of sung vowels ([ a], [e ], C i ]) produced by ten baritone/bass singers at the University of North Texas was analyzed by the use of Fourier analysis and electronic digital equipment. This procedure and equipment produced results over a wider frequency range with greater accuracy than prior studies on vocal timbre. The study sought to answer the following questions: (1) Using formant regions between 0-20 kHz for comparison, what similarities and differences can be observed among spectra produced from [a], [e], and [ i ] vowels sung by baritone/bass singers? (2) Using formant regions between 0-20 kHz for comparison, what similarities and differences can be observed among spectra produced from [a], [ e ], and [ i ] vowels sung by baritone/bass singers with regard to individual singers? (3) Approximately what vocal-tract tube lengths were used by baritone/bass singers when performing [a], [e ], and [ i ] vowels? (4) What similarities in vocal-tract tube lengths can be generalized as to [ a], L e ], and [ i ] vowels sung by individual baritone/bass singers? The results of the study suggested that: (1) Below approximately 4 kHz formant frequency location can be generalized by a specific vowel between subjects. (2) Above 4 kHz the generalization of formant frequency location is difficult between subjects singing the same vowels, but general frequency location for formants can be identified between samples produced by the same singer performing different vowels. (3) Subjects did alter their vocal-tract lengths as different vowel sounds were performed, but no overall pattern of tube length with reference to specific vowels was indicated. (4) Each singer did use a unique vocal-tract length. The average estimated vocal-tract length for all subjects singing a specific vowel revealed little variation in tube length with an average tube length of 278.97 mm.

Identiferoai:union.ndltd.org:unt.edu/info:ark/67531/metadc330963
Date08 1900
CreatorsTolin, Craig Edmond
ContributorsMay, William V., Rainbow, Edward L., Baird, Edward A., Pinizzotto, Russell F.
PublisherUniversity of North Texas
Source SetsUniversity of North Texas
LanguageEnglish
Detected LanguageEnglish
TypeThesis or Dissertation
Formatviii, 98 leaves: ill., Text
RightsPublic, Tolin, Craig Edmond, Copyright, Copyright is held by the author, unless otherwise noted. All rights reserved.

Page generated in 0.002 seconds