Adaptive control and patterns of stability of consonant production:
General Material Designation
[Thesis]
First Statement of Responsibility
B.-e. E. Shia
Title Proper by Another Author
An acoustic study of English stop consonants
Subsequent Statement of Responsibility
H. L. Sussman, Bjorn
.PUBLICATION, DISTRIBUTION, ETC
Name of Publisher, Distributor, etc.
The University of Texas at Austin
Date of Publication, Distribution, etc.
1995
PHYSICAL DESCRIPTION
Specific Material Designation and Extent of Item
194
DISSERTATION (THESIS) NOTE
Dissertation or thesis details and type of degree
Ph.D.
Body granting the degree
The University of Texas at Austin
Text preceding or following the note
1995
SUMMARY OR ABSTRACT
Text of Note
This study presents acoustic data on the impact of speaking style, vowel context, closure duration and aspiration on specific production characteristics (closure duration and F2 locus frequency) of a given target stop consonant. The stop consonants (/b, d, g, p, t, k/) investigated were embedded within a VCV2 structure, of which V2 cover a wide range of English vowels. The acoustic data in this study show that speech signals are often shaped by physiological, linguistic and social demands. It was found that F2 onset frequencies, which acoustically denote the place of articulation of stop consonants, were sensitive to speaking style. In comparison to normal speech, clear speech was characterized by a slightly steeper slope and lower y intercept when F2 onset frequency was plotted as a function of F2 mid of vowels in the (h-d) context (F2 target). The close interaction between formant frequency and duration (duration-dependent undershoot) reported for vowels (Lindblom, 1963; Moon & Lindblom, 1994) was not observed in stop consonants, which appeared to be resistant to the influence of closure duration. Stop consonants exhibited a slight effect due to preceding vowel contexts. The frequency of F2 onset tended to be greater when the initial vowel was a front vowel (/i/) than a mid (/usd\partialusd/) or a back vowel (/u/). As for the effect of aspiration, voiced stops were found to be less coarticulated with the following vowels than their voiceless cognates when F2 onset was taken at the CV boundary; when F2 onset measurements were taken at the earliest presence of F2 energy within the burst, voiceless stops were less coarticulated with the following vowels than their voiced cognates. These differences can be explained by the timing of F2 onset vis-a-vis the full extent of F2 transition.