The relation between focus effects in production and exemplar locations in perception for stop types in English, Japanese, and Korean
Eric Oglesbee, Kenneth Alan De Jong · The Journal of the Acoustical Society of America · 2007
This paper examines focus-induced variability in the production of labial stop contrasts in English, Japanese, and Korean, pursuing differences between the languages in how voice onset time (VOT) and fundamental frequency (F0) are used to indicate differences between stop types. Language differences are compared to differences in the perceptual usage of these acoustic cues using AMBEL, an adaptive tracking algorithm that allows subjects to search multi-dimensional stimulus arrays for best exemplars. All three languages exhibit category differences in VOT, though Japanese and Korean exhibit overlap for two of the categories. As found previously, for these categories, both Japanese and Korean exhibit large differences in F0 in the following vowel. Except for the smaller F0 differences in English, all of these VOT and F0 differences are systematically expanded in a focus condition. In particular, VOT increases for the more aspirated stops in all three languages, and substantial rising F0 patterns are observed in lenis Korean and zero-lag Japanese stops. Perceptual data match these focus effects; best exemplars differ in VOT in all three languages, and in Japanese and Korean, where substantial F0 focus effects were observed, F0 differences affect the identification of best exemplars. [Work supported by NSF.]