Voice quality and pitch contrast in non-native Korean

Abstract

In this study, I compare the effects of linguistic experience on voice quality (H1-H2) and fundamental frequency (f0) in Korean stops among native and non-native Korean speakers. Native speakers of Chinese, English, Korean, and Spanish produced Korean words in a /CVC/ context, and H1-H2 and f0 of the initial stops in each set of materials were measured. Korean and Chinese speakers showed creakiness (smaller H1-H2) for Korean tense stops and breathiness (larger H1-H2) for lenis and aspirated stops, whereas English and Spanish speakers showed relatively larger H1-H2 for all stops. For f0 values, Korean and Chinese speakers displayed a lower f0 for lenis, an intermediate f0 for tense, and a higher f0 for aspirated stops. For English speakers, however, lenis and tense stops were merged in the lower f0 region, and aspirated stops showed a higher f0. In Spanish speakers, tense and aspirated stops merged in the higher f0 region, and lenis stops showed a lower f0. These results demonstrate a strong effect of linguistic experience on voice quality and f0: speakers of Chinese were more accurate in replicating Korean stops than were speakers of English or Spanish, languages that lack phonemic voice quality and tone contrasts.