Abstract
Speech samples from 40 hearing impaired children who ranged in age from 6 to 11 were recorded, using sentence intonation and word accent as test materials. The average hearing levels of the subjects ranged from 50 to 130dB. Changes in the fundamental frequency of the speech samples were observed acoustically, and differences in tone changes between declarative and interrogative sentences, and those between different positions of the accent kernel of words, were analyzed. At the same time, the quality of the utterances was judged perceptually, and they were classified into three categories; correct, wrong, and in-between. It was found that the results of the perceptual classification could be explained from the results of the acoustical analysis, if the region of correct utterances was set according to the characteristics of standard utterances spoken by normal subjects. This implies that the qualities of intonation and accent can be evaluated objectively by acoustical analysis. The relationship between the results obtained from these methods of evaluation and the characteristics of hearing loss were examined. It was suggested that control of voice pitch in the utterance is closely related to hearing level in the low frequency range, and that ability to control voice pitch can be predicted by from the hearing levels at 250 Hz and 500 Hz.
