Articulatory-acoustic relations in Cantonese vowel production

Wai-Sum Lee*

*Corresponding author for this work

    Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

    Abstract

    The study evaluates the quantal theory (Stevens, 1972, 1989) through investigating the articulatory-acoustic relations in two sets of Cantonese point vowels, the long [i: u:] and the medium-long [i u]. It determines the sensitivity of vowel formant frequencies to variations in linguo-palatal constriction location and constriction size along the vocal tract length. The articulatory and acoustic data on the vowels obtained in this study show non-linear relations between the acoustic consequences and the up-down and front-back displacements in linguo-palatal constriction, as what the quantal theory predicts. Moderate displacements in constriction location and constriction size bring about large perturbation in acoustic parameter. The level of sensitivity of formant frequencies to variation in tongue displacement is lower in [i: u:] than [i u], suggesting that the quantal nature is more apparent in [i: u:] than [i u]. For the Cantonese point vowels, F1 is more sensitive to variation in constriction location than constriction size, whereas F2 is more sensitive to variation in constriction size than constriction location. A difference between Cantonese and English is that, in English, both F1 and F2 are relatively more sensitive to variation in constriction size than constriction location. © 2012 Acoustical Society of America.
    Original languageEnglish
    Article number060002
    JournalProceedings of Meetings on Acoustics
    Volume15
    DOIs
    Publication statusPublished - May 2012
    EventAcoustics 2012 and the 163rd Meeting of the Acoustical Society of America - Hong Kong, China
    Duration: 13 May 201218 May 2012
    https://acousticalsociety.org/program-acoustics-2012-hong-kong/

    Fingerprint

    Dive into the research topics of 'Articulatory-acoustic relations in Cantonese vowel production'. Together they form a unique fingerprint.

    Cite this