Effect of linguistic background on convergence of prosodic rhythm

Abstract

Speech patterns of the interlocutors become more similar to each other over the course of an interaction. These spontaneous speech adaptations, or phonetic convergence (PC), have been demonstrated for segmental features, such as vowels and voice onset times (VOT) and for suprasegmental features, such as stress. In this study, speaker adaptations to speech rhythm are examined before and after an interactive map task. Using American English and Indian English speakers, convergence was measured using the centroid of the envelope modulation spectrum (EMS + centroid, Rao & Smiljanic, 2011). This spectral measure of rhythm goes beyond considering consonantal and vocalic duration variability, as used in the traditional rhythm measures, and includes information about syllable prominence, stressed and unstressed syllable variation and distribution, and pauses and disfluencies. This research will allow us to examine whether language background has an effect on convergence of global speech properties, such as linguistic rhythm. The results of this study add to our current knowledge of features that are subject to imitation in the speech of dialogue partners.