Comparison of fundamental frequency in normal-hearing children and children with various severities of hearing loss

Hearing is considered as an important sense in closed-loop control and is one of the crucial factors affecting voice production due to its role in providing the needed feedback for voice control. In the absence of auditory feedback، as in individuals with hearing loss، the motor control for the muscles associated with phonation and speech production is negatively affected. As a result، they are not able to regulate their fundamental frequency، which in return leads to impaired speech intelligibility. The purpose of this study was to compare the acoustic parameter F0 in children with different degrees of hearing loss with the same parameter in their normal peers. The voice samples of 28 children with hearing loss (moderate، moderate to severe، severe and profound) and 28 normal hearing children with the age range of 7 to 9 years were recorded. All participants were monolingual speakers of Farsi. The examinees were asked to produce the vowel /a/ for 3 times. The average F0 was calculated for each examinee using PRAAT software (version 5. 3. 13). One-way ANOVA test was conducted to examine the differences in F0 values across different groups and the post-hoc Dunnett’s test was employed for evaluating the statistical significance of the differences observed. The highest and lowest values of F0 belonged to the subjects with profound hearing loss and those with normal hearing respectively. Results of ANOVA test revealed that the mean values of F0 were different in the studied groups (P = 0. 025) and these differences were statistically significant as children with profound hearing loss and the control group were compared (P = 0. 005). The mean value of F0 increases as the severity of hearing loss increase. This condition might be accounted for by a mechanism for compensating the lack of auditory feedback. The findings of the current research indicate that the more severe the hearing loss، the more average F0 would be expected; but further research is required to delineate a cut-off point between the normal and disease-induced voice samples.