Acoustic and Perceptual Description of Vowels in a Speaker With Congenital Aglossia

Abstract

The goals of this study were to (a) compare the vowel space produced by a person with congenital aglossia (PWCA) with a typical vowel space; (b) investigate listeners’ intelligibility for single vowels produced by the PWCA, with and without visual information; and (c) determine whether there is a correlation between scores of speech intelligibility of PWCA speech and the acoustic properties of those speech samples. The main objective of this study was to determine whether a PWCA was able to compensate for the lack of tongue and whether listeners were able to compensate perceptually for the possible atypical acoustics of the PWCA. Cineradiography for this article was limited to observation of gross function of the tongue base and mylohyoid. An audiovisual recording of the PWCA speaker’s output was obtained for a series of isolated vowels, diphthongized vowels, and vowels in monosyllables. Production of vowels was analyzed acoustically and perceptually. Vowels were presented to listeners under two conditions: audiovisual and audio only. Paired differences sample tests revealed no statistical differences in intelligibility for the audio versus audiovisual conditions. Mean intelligibility for vowels was 78.5% overall. Intelligibility was a function of vowel position, with the front vowels revealing the least intelligibility and the back vowels revealing the greatest intelligibility. Quantitative analysis of F1–F2 formant data revealed that the speaker’s front vowels showed greater distances from the back vowels when compared with the F1–F2 means of normative data.