Glottalization of word-initial vowels as a function of prosodic structure

Abstract

Earlier work on the glottalization of word-initial vowels sought an account in terms of the morphosyntactic hierarchy and isolated facts about stress, without accounting for the possible role of phrase-level prosodic structure. More recent work based on prosodic theory (Pierrehumbert & Talkin, 1992; Pierrehumbert, 1995) has shown that prosodic constituents and prominence are important factors in the description of glottalization: word-initial vowels glottalize with higher frequency at the beginnings of intonational phrases, and to a greater degree if the word is pitch-accented. The work reported here extends Pierrehumbert & Talkin’s findings using a different method, and reports additional dependencies on prosodic structure. The study is based on a different style of speech and a wider variety of prosodic and segmental contexts, in a speech corpus that was produced with communicative intent (FM radio news style, 3709 word-initial vowels produced by five speakers, three female and two male). Analysis of this corpus shows that glottalization of word-initial vowels is influenced by fullvs.intermediate intonational phrase boundaries and pitch accent on the target syllable or word, but the effect of the preceding segmental context is small in comparison. These results are robust despite substantial interspeaker differences in both the distribution and nature of glottal onsets, and illustrate the value of revisiting earlier morphosyntactic analyses with prosodic structure in mind.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 467 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.