LINGUIST List 33.3433
Fri Nov 04 2022
Diss: Phonetics; Feda Negesse: '' Vowel And Voice Characteristics Of Prelingually Deaf Children: Acous Tic And Perceptual Experiments''
Editor for this issue: Sarah Goldfinch <sgoldfinchlinguistlist.org>
Date: 29-Oct-2022
From: Feda Negesse <fedan2010
gmail.com>
Subject: Vowel And Voice Characteristics Of Prelingually Deaf Children: Acous Tic And Perceptual Experiments
E-mail this message to a friend Institution: Addis Ababa University
Program: Experimental phonetics
Dissertation Status: Completed
Degree Date: 2013
Author: Feda Negesse
Dissertation Title: Vowel And Voice Characteristics Of Prelingually Deaf
Children: Acous Tic And Perceptual Experiments
Linguistic Field(s): Phonetics
Dissertation Director(s):
Valerie Hzan
Lacerda Francisco
Dissertation Abstract:
This study was intended to examine the effects of prolonged auditory deprivation on the vowel and voice production characteristics of children who received no intervention. (1) It was expected that the deaf children in the study would significantly differ from the norms with respect to acoustic characteristics of their vowels and voices due to (i) the deterioration of speech motor programs acquired before the onset of deafness, (ii) lack of well-tuned feedforward commands and (iii) the absence of auditory feedback to guide their vowel and voice productions. Yet, (2) it was expected that the children would be able to maintain some aspects of acoustic characteristics of their vowels and voices because of (i) the availability of somatosensory feedback, (ii) the visibility of articulations for some vowels, and (iii) the resistance of vowels to deafness. Finally, the vowels of the deaf children were hypothesized to be more difficult to identify for naive listeners than for experienced listeners because of a lack of familiarity with the speech of deaf people.
Acoustic and perceptual experiments were carried out to test the hypotheses formulated above. The acoustic data for the experiments were recorded from six hearing children and six prelingually deaf children who acquired Amharic as their first language and volunteered to participate in the study. The required acoustic features were extracted and processed with different speech and statistical software packages. The spectral and temporal features extracted for the vowels included F1, F2, F3, FO, and vowel duration while the voice parameters consisted of FO, jitter, shimmer, and other spectral features. The perceptual experiment involved 30 participants, 15 of whom had no experience with the speech of deaf people and 15 of whom had 3-4 years of familiarity with the speech of hearing-impaired people.
The acoustic experiment partly confirmed the hypothesis that spectral features of vowels produced by the experimental group would differ from those of the comparison group. A significant increase of F1, heavy centralization of F2, and longer duration of vowels were observed for the experimental group. Regarding FO and F3, the experimental group did not significantly differ from the comparison group. Vowels of deaf children appear to be more variable, less differentiated, and poorly classified and identified. However, it was noted that vowels with visible articulations appear to be less affected and that vowels maintained their collective features such as frontness and backness. Furthermore, significant differences were found between the voice production characteristics of the experimental and the comparison groups. The voice FO, jitter, shimmer, and maximum phonation time were significantly greater in the deaf children in agreement with the findings of previous research.
Finally, the hypothesis that naïve listeners would find the vowels of deaf children more difficult to identify than familiar listeners was not supported by the perceptual experiment; probably due to the loss of essential perceptual cues, both groups might have experienced similar difficulty in identifying the vowels. Yet, familiar listeners had shorter response times and more favorably rated the quality of vowels of the deaf children but the differences failed to reach statistical significance.
In general, the speech motor commands acquired prior to the onset of deafness appears to decay, and somatosensory feedback alone could not seem to help the children acquire and maintain normal vowel and voice productions. Future studies should focus on investigating how the absence of auditory feedback affects each acoustic characteristic of vowels and voices of children. It is also important to examine the relative effects of deterioration of speech motor programs and feedforward motor commands on vowel and voice qualities of deaf children.
Page Updated: 04-Nov-2022