Stochastic Suprasegmentals: Relationships between Redundancy, Prosodic Structure and Syllabic Duration

Matthew P. Aylett

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

Abstract

Prosodic Prominence and prosodic boundaries have been shown to effect syllabic durations. However another factor, redundancy, also appears to have a major impact. More common words and words you can easily predict from context (more redundant) tend to be articulated less clearly and so also have a tendency to have shortened syllabic durations.
This paper explores the relationship between measures of redundancy, prosodic structure and syllabic duration of a large corpus of spontaneous speech. Although 50% (r=0.71) of syllable variation is predictable from measures of accentedness, break index and other prosodic parameters, word frequency alone predicts 11% of the duration variation. Combining prosodic information and redundancy measurements improves prediction by 0.75% (r=0.72), suggesting that although redundancy measurements can offer a statistically independent contribution to predicting syllabic duration, prosodic structure implicitly represents most of the variation caused by redundancy.
Original languageEnglish
Title of host publicationProceedings of the XIVth International Congress of Phonetic Sciences
Subtitle of host publicationICPhS-14
EditorsJohn J. Ohala, Yoko Hasegawa, Manjari Ohala, Daniel Granville, Ashlee C. Bailey
PublisherAmerican Institute of Physics
Pages289-292
Number of pages4
ISBN (Print)1-56396-898-3
Publication statusPublished - 1999
Event14th International Congress of Phonetic Sciences - San Francisco, United States
Duration: 1 Aug 19997 Aug 1999

Conference

Conference14th International Congress of Phonetic Sciences
Country/TerritoryUnited States
CitySan Francisco
Period1/08/997/08/99

Fingerprint

Dive into the research topics of 'Stochastic Suprasegmentals: Relationships between Redundancy, Prosodic Structure and Syllabic Duration'. Together they form a unique fingerprint.

Cite this