فاطمه نعمتی

Cochlear implant provides access to auditory feedback for sever to profound sensory-neural hearing impaired patients and has been reported to result in considerable language development among them. However, comparing with Normal hearing speakers, falling short of demonstrating the same speech characteristics. Some of these differences relate to the production of vowels which are determinant segments in speech intelligibility. The purpose of this study is to compare the acoustic features of Persian vowels produced by CI and NH speakers. 10 CI children with at least 3 years of implantation and 10 normal hearing children entered into the study by convenience sampling. They repeated 1920 CV syllables made up of all Persian plosives and vowels, after a speech therapist. As the objective representations of the accuracy of vowel articulation and the main acoustic cues in auditory perception for listeners, first and second formants of vowels (F1 and F2) of CI and NH children, were analyzed and compared with each other. Due to nonparametric distribution, Mann-Whitney test was used to compare the differences between cases and controls. In all cases, significance level was set at 5%. The comparison of F1 and F2 of six Persian vowels produced by CI and NH children demonstrated statistically significant differences in eight cases. CI children formed all vowels except for /?/ more open than normal, and 3 back vowels more front. The shift to more open vowels could be traced back to the training method. As the trainers were observed to articulate the vowels exaggeratedly with open mouth to give visual cues for their place of articulation, the vowels by CI children had been formed more open than normal vowels. Regarding F2 of vowels, CI children produce the back ones more front which could be traced to the lack of visual cues, while the F2 of front vowels are normal. The difference between F2 of vowels by CI children suggests that time span after the surgery might have benefited CI