Abstract
Differences in time structures of voiced and unvoiced stop consonants contained in Japanese monosyllables were investigated. Perceptual tests were conducted with several modified stop consonants formed by partially cutting and rearranging tape recordings of stop consonants. Conclusion obtained from the results of the tests is as follows: Primary cues to distinguish |b|, |d|, |g| from |p|, |t|, |k| are (1) presence of preceding buzz bar, (2) rising characteristic of pitch frequency of adjacent vowel, (3) transition of first formant, and secondary cues are weaker burst intensity and absence of aspiration.
