EFFECTS OF TALKER DIFFERENCES ON SPEECH INTELLIGIBILITY IN THE HEARING IMPAIRED.
Item
-
Title
-
EFFECTS OF TALKER DIFFERENCES ON SPEECH INTELLIGIBILITY IN THE HEARING IMPAIRED.
-
Identifier
-
AAI8103947
-
identifier
-
8103947
-
Creator
-
MARGULIES, MARIS KABAT.
-
Contributor
-
Harry Levitt | Dr. Thomas Fay
-
Date
-
1980
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Health Sciences, Audiology
-
Abstract
-
Five groups of ten listeners each (normal hearing, condutively impaired, flat-, moderately sloping-, and severely sloping sensori-neurally impaired) were presented with recordings of the CID sentences of everyday speech (CHABA) spoken by ten different talkers (five male and five female). In addition, recordings of nonsense syllables for two talkers (one male and one female from the above group) were also used. Five types of background noise were added to the sentence materials: high-pass, low-pass, white, male-spectrum and female-spectrum shaped noise. The ratio of speech peaks to rms noise level was set at 0 dB for the sentence materials. Four experimental conditions were used for the nonsense syllable materials: frequency filtering with a slope of 0 dB/octave, +6 dB/octave and -6 dB/octave (all in quiet) and 0 dB/octave slope condition in noise at a S/N ratio of 0 dB. In addition to obtaining intelligibility scores, subjects were required to rate the intelligibility of each talker on a scale of 1-7.;Significant differences in intelligibility were obtained between talker gender, experimental groups, experimental conditions (added background noise and/or frequency filtering), and listener age. On the sentence materials the female talkers, on the average, were more intelligible than the male (for a common speech-to-noise ratio). There was, however, a substantial overlap between the intelligibility scores of individual male and female talkers. For the nonsense syllable materials, where only one talker of each gender was used, the male talker yielded higher discrimination scores than the female. This male talker happened to be the most intelligible of the male talkers used in the study, whereas the female talker was of average intelligibility. No significant differences were observed between male and female listeners.;As expected, intelligibility scores decreased with increase in degree of hearing impairment but the relative masking and filtering effects varied between listener groups and talker gender. Younger listeners scored higher than older listeners except in the conductive hearing loss group. This apparent exception is believed to be the result of a very high score shown by one listener.;The intelligibility rankings showed a higher degree of consistency with the objective measure of intelligibility for the normal hearing listeners. The consistency between the rankings and measured intelligibility decreased as the degree of hearing impairment, especially in the higher frequencies, increased. There was no significant male-female talker difference in the consistency of the ratings.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
-
Program
-
Hearing and Speech Sciences
