This study examined how Korean learners of English produce lexical and phrasal stress compared to native English speakers. Forty Korean speakers learning English as a foreign language and 11 native English speakers read aloud sentences designed to elicit contrasts in lexical stress (e.g., súspect as a noun vs. suspéct as a verb) and phrasal stress (príntout as a compound noun vs. print óut as a phrasal verb). Acoustic analysis on the recorded tokens (maximum pitch, maximum intensity, and duration of words and syllables) revealed that the English speakers used duration as the most reliable cue for both lexical and phrasal stress. Intensity was used as another significant cue to lexical stress but not to phrasal stress, and no significant use of F0 was found for either stress type. Korean speakers with higher English proficiency marked the lexical stress contrast more clearly than those with lower English proficiency, mainly by producing the first syllable in nouns longer than in verbs. However, the Korean speakers did not employ any of the three acoustic correlates examined to mark the contrast in phrasal stress, regardless of their proficiency in English. These results are interpreted with relevance to previous findings in the literature, and suggestions are made regarding relevant teaching methods.
