Perception of the strength of prosodic breaks in three conditions: Explicit pause, implicit pause, and no pause

In this study we examine the perceptual strength of prosodic boundaries in Hebrew speech. The stimuli consisted of 28 sequences of two inter-pausal units (IPUs) taken from the Map Task recordings in Hebrew. Listeners were exposed only to the silent pause following the first IPU (hence, Explicit pauses) while the second pause was omitted (hence, Implicit pauses) thus creating a stimulus model of IPU-pause-IPU. Ten female listeners labeled the strength of each break between adjacent words on a scale from 1 (no break) to 5 (strong break). Higher average scores were assigned to the implicit pauses as compared to the explicit ones, however scores for explicit pauses received higher agreement between raters. Moreover, we found only borderline significant influence of the explicit pause duration on the raters' scores. Looking at gender differences, the results suggest that raters' scores were higher when the speakers were females. Further, an interaction was found between the gender of the speaker and the gender of the recipient (i.e., the interlocutor). In particular, female speakers received a higher score overall, and for male speakers the rating was higher when they spoke to males than to females.