Acoustic correlates of English rhythmic patterns for American versus Japanese speakers

This study investigates acoustic correlates of English rhythmic patterns for 20 American English speakers (AS) and 42 Japanese learners of English (JS). The results indicate that for AS in an English sentence where monosyllabic content and function words alternate, the vowels in content words are over twice as long as those in function words, resulting in alternating long-short vowels. In contrast, the JS show no stress-related duration control and realize a similar rhythmic pattern mostly through recursive high-low fundamental frequency (F0). In a sentence with a sequence of content words in which 4 stressed syllables occur successively, the AS show recursion of strong-weak syllables by means of F0, intensity and first formant, whereas JS show inconsistent stress patterns. These results indicate that the AS apply different strategies for implementing rhythmic alternation depending on sentence stress patterns, and these strategies are different from those of JS.