1999The Journal of the Acoustical Society of AmericaRequires access

Acoustic evidence for dynamic formant trajectories in Australian English vowels

Catherine Watson, Jonathan Harrington

Open publisher page 173 citations

Abstract

The extent to which it is necessary to model the dynamic behavior of vowel formants to enable vowel separation has been the subject of debate in recent years. To investigate this issue, a study has been made on the vowels of 132 Australian English speakers (male and female). The degree of vowel separation from the formant values at the target was contrasted to that from modeling the formant contour with discrete cosine transform coefficients. The findings are that, although it is necessary to model the formant contour to separate out the diphthongs, the formant values at the target, plus vowel duration are sufficient to separate out the monophthongs. However, further analysis revealed that there are formant contour differences which benefit the within-class separation of the tense/lax monophthong pairs.

About this research paper

What this paper is about

The extent to which it is necessary to model the dynamic behavior of vowel formants to enable vowel separation has been the subject of debate in recent years. To investigate this issue, a study has been made on the vowels of 132 Australian English speakers (male and female). The degree of vowel separation from the formant values at the target was contrasted to that from modeling the formant contour with discrete cosine transform coefficients. The findings are that, although it is necessary to model the formant contour to separate out the diphthongs, the formant values at the target, plus vowel duration are sufficient to separate out the monophthongs. However, further analysis revealed that there are formant contour differences which benefit the within-class separation of the tense/lax monophthong pairs.

Why it matters

OpenAlex reports 173 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

The extent to which it is necessary to model the dynamic behavior of vowel formants to enable vowel separation has been the subject of debate in recent years. To investigate this issue, a study has been made on the vowels of 132 Australian English speakers (male and female). The degree of vowel separation from the formant values at the target was contrasted to that from modeling the formant contour with discrete cosine transform coefficients. The findings are that, although it is necessary to model the formant contour to separate out the diphthongs, the formant values at the target, plus vowel duration are sufficient to separate out the monophthongs. However, further analysis revealed that there are formant contour differences which benefit the within-class separation of the tense/lax monophthong pairs.

Key concepts: Formant, Diphthong, Vowel, Acoustics, Mathematics, Speech recognition, Computer science, Physics

Related papers

Back to paper searchBrowse research topicsOriginal source
Acoustic evidence for dynamic formant trajectories in Australian English vowels — Research Paper | ScholarLens