The effect of talkers' language dominance on subjects' speech production of sibilant fricatives

Abstract

This study investigates the effect of talkers' language dominance on subjects' sibilant production in a bilingual community. Guoyu (Taiwanese Mandarin) has 3 sibilants: alveolar /s/, retroflex /s/ and alveolo-palatal /ɕ/, while Taiwanese (a Southern Min dialect) only has /s/, which is palatalized before /i/. Previous studies have shown that Taiwanese-dominant speakers in Taiwan has a merged category of /s/, /s/ and /ɕ/. In addition, they treat [s] and [ɕ] as allophones of /s/. On the other hand, Guoyu-dominant speakers have a more distinctive three-way contrast of sibilants. This study explores whether listening to talkers with different language dominances affects subjects' speech production of Guoyu sibilants. Two female talkers, one is Taiwanese-dominant and the other is Guoyu-dominant, recorded Guoyu words containing target sibilants in word-initial position with comparable vowels. Forty bilingual adults' productions were elicited in a repetition task blocked by talker. The spectral centroid is obtained from the middle 40ms of each sibilant, along with the onset F2 of the following vowel. The two acoustic measures were plotted against each other and separated by talker. Preliminary results show that these subjects' productions differ when prompted by different talkers. Additional statistical tests will be performed to explore this production difference.