Date of Award
2006
Degree Type
Thesis
Degree Name
Master of Science
Program
Communication Sciences and Disorders
Supervisor
Dr. Philip C. Doyle
Abstract
The purpose of the present investigation was primarily focused on how naive listeners assess perceptual dimensions underlying voice quality (VQ) for dysphonie and normal voice samples over time. Specifically, this investigation evaluated whether a listener’s auditory perception of voice changed with repeated, systematic exposure to voice samples. In doing so, this study was concerned with the consistency of listener judgments. Thirty-two naïve, young, normal hearing listeners were asked to rate dysphonie (24 samples) and normal voice stimuli (6 samples) on five perceptual dimensions: overall severity, breathiness, roughness, strain, and pitch. Randomized samples were rated over four individual listening sessions, each separated by approximately one week. Perceptual assessment of the above five dimensions were based on the procedural methods of the Consensus Auditory Perceptual Evaluation - Voice (CAPE-V) provided by the American Speech-Language Hearing Association (ASHA, 2005). However, all ratings were based on judgments of the second sentence of the Rainbow Passage. Listeners rated CAPE-V dimensions for each of the 30 experimental stimuli. Perceptual evaluation was performed using a 100mm visual analogue (VA) scale for each of the five dimensions. The perceptual data from the VQ ratings were then used to determine whether listeners were consistent in their auditory-perceptual judgments of voice quality in response to repeated, systematic exposure to voice samples. That is, did individual listener perception change with repeated, systematic exposure to voice samples? iii Statistical analyses using interclass correlations (ICC) and pair sample T- test revealed first, that the dimensions of overall severity and strain were consistently rated by naïve listeners while roughness, breathiness, and pitch achieved low consistency (ICC) scores. Second, listener perception for the dimensions of overall severity and roughness remained stable despite repeated systematic exposure to the voice stimuli, but auditory perceptual judgments for the dimensions of breathiness, strain, and pitch showed evidence of significant change over the four sessions (comparison of ratings between Session 1 and Session 4 for each dimension). The above results are reflective of: a) the naive listener’s perceptual stability with reference to a normal voice standard, and b) a potential change in the naive listener’s internal prototype due to repeated exposure, and hence, increased familiarity with the stimuli. These prototypes formed the basis for the auditory perceptual judgments. Based on these data, naive listeners appeared able to consistently rate broad-based dimensions of voice quality such as overall severity while more specific, isolated dimensions such as roughness, breathiness, strain, and pitch were inconsistently rated. Also, repeated exposure to the stimuli resulted in a significant change in auditory perceptual judgments for the dimensions of breathiness, strain, and pitch, while ratings for overall severity and roughness remained stable. These results have both theoretically importance, as well as clinically value specific to the auditory-perceptual evaluation of voice.
Recommended Citation
Bulfon, Andria, "EVALUATION OF LISTENER CONSISTENCY FOR PSYCHOPHYSICAL RATINGS OF VOICE QUALITY" (2006). Digitized Theses. 5079.
https://ir.lib.uwo.ca/digitizedtheses/5079
