Perception of French stops by French and American English speakers

Abstract

This study examined how native French and native American English speakers perceive edited and unedited word-initial stops produced by two monolingual Parisian speakers. Most words used in this study exist in both languages, yet are produced differently. Listeners were asked to identify the initial stops of French monosyllabic words recorded in isolation. The words were unedited (condition A), edited (condition B), and whispered (condition C). In condition A, words were produced in isolation at modal vocal intensity. In condition B, voicing lead was removed from words beginning with /b, d, g/, and in condition C, words were naturally whispered. Participants were asked to listen to each stimulus and then identify the initial consonant (b, d, g, p, t, k). Statistical analyses show that native French speakers performed significantly better than native American English speakers on identification of the word initial consonant in condition A (unedited words spoken at modal vocal intensity). In condition B (voicing lead removed from words produced at modal vocal intensity), native American English speakers performed significantly better than the native French speakers. Finally, both language groups were equally impaired in condition C (whispered speech). Errors were analyzed as a function of the listener’s language phonetic inventory.