By the same authors

Information structure, affect, and prenuclear prominence in American English

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Full text download(s)

Published copy (DOI)

Author(s)

Department/unit(s)

Publication details

Title of host publicationProceedings of INTERSPEECH 2018
DatePublished - 2 Sep 2018
Pages1848-1852
Number of pages5
Original languageEnglish

Abstract

The influence of information structure (IS: givenness, accessibility, newness and focus) on pitch accent assignment and acoustic prominence measures of prenuclear words was investigated for American English speech elicited through read production of mini-stories. Results showed a consistent pattern of accenting the initial content word in the sentence, supporting an analysis of prenuclear accent as structural, or ‘rhythmic’. While no association was observed between IS and accent type (e.g., H*, L*, L+H*, L*+H), the acoustic-phonetic realization of prominence was modulated by information structure. In particular, words that carry contrastive focus generally showed more extreme f0 excursions relative to the average. In addition, there was a strong influence of speaking style or ‘affect’ on both pitch accent type and the acoustic-phonetic realization of prominence. Speakers were more likely to produce L+H* accents in a lively than a neutral speaking style. Differences in affect were also strongly reflected in f0 excursion, duration, and amplitude within the target word. Overall, this study indicates both linguistic (information structure) and paralinguistic (affect) influences on the phonetic implementation of prenuclear prominence, with varying influence of these two factors on the phonological assignment of prenuclear pitch accents.

Bibliographical note

This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy. Further copying may not be permitted; contact the publisher for details.

Discover related content

Find related publications, people, projects, datasets and more using interactive charts.

View graph of relations