歯擦音の母音無声化・VOT分布・促音挿入

要旨

The distribution of voice onset time (VOT) in voiced plosives usually has bimodal feature in Japanese.One of plausible explanation for "two peaks" is that the command of vocal fold vibration is controlled on two different reference points: one is based on the release timing of closure, the other reference point is the phonological boundary which indicates both the ending timing of the preceding consonant and the starting timing of the following vowel.This paper proposed that the consonant-vowel boundary has relation with various phonological phenomena in Japanese: vowel devoicing, voicedness of obstruents, geminate insertion/inhibition in Japanese loanwords and so on.Finally this paper shows that the model according to consonant-vowel boundary can simulate the distribution of VOT.

主題