An Acoustic Analysis of Stressed and Unstressed Vowels in English Nouns

This paper investigates the acoustic properties of English stressed and unstressed vowels by conducting production experiments, in which Korean, Japanese, and English speakers were asked to concatenate base words and their matching suffixes (e.g., úniform + ity--＞ùnifórmity). Specifically, the ratios of stressed to unstressed vowels were measured for duration, fundamental frequency, and intensity. The results showed that language backgrounds of the participants played an important role with respect to duration as the ratio of stressed to unstressed vowels for duration was much longer for the native English speakers than for the Korean and Japanese speakers. On the contrary, the ratio of stressed to unstressed vowels for intensity was greater for the Japanese and Korean speakers than for the English speakers. However, no significant ratio difference for fundamental frequency was found between stressed and unstressed vowels among the three language groups. Thus, it is suggested that Korean and Japanese speakers were not similar to English speakers in implementing the difference between stressed and unstressed vowels with respect to duration and intensity. It is also shown that suffix types played a significant role with respect to duration as the participants across language backgrounds showed higher duration ratios of stressed to unstressed vowels before stress shifting suffixes (-ity, -ation) than before stress non-shifting suffixes (-ness, -ment). Yet, there was no main effect of suffix type irrespective of stress shifting or non-shifting suffixes with regards to fundamental frequency and intensity. Also, the patterns of suffix types varied depending on individual suffixes and language groups. Implications of the findings for English stressed and unstressed vowels were further discussed.