Abstract
Measurements were made of the correspondence between acoustic (declination) units and auditory (intonation) units in American English conversation in an attempt to relate these two levels of description. The results showed that 99% of the acoustic unit boundaries coincided with auditory unit boundaries, indicating that the acoustic cues of F0 reset, pause, and speaker change used in defining the acoustic units were perceived, verifying the validity of both types of analyses. However, the units differed in scope, with many more auditory units occurring than acoustic units. This result suggests that the common practice of assuming the ‘intonational phrase’ to be the primary domain of prosodic phenomena such as declination does not adequately account for our data of larger periods of speech. Both acoustic and auditory analysis are needed for a satisfactory description of intonation. The establishment of a methodological connecting link between the acoustic and auditory dimensions suggests that further comparisons of such different data types may yield guidelines for regularizing the relationship between the two levels.
Get full access to this article
View all access options for this article.
