Vowel and consonant contributions to vocal tract shape.

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

The purpose of this study was to develop a method by which a vowel-consonant-vowel (VCV) utterance based on x-ray microbeam articulatory data could be separated into a vowel-to-vowel transition and a consonant superposition function. The result is a model that represents a vowel sequence as a time-dependent perturbation of the neutral vocal tract shape governed by coefficients of canonical deformation patterns. Consonants were modeled as superposition functions that can force specific portions of the vocal tract shape to be constricted or expanded, over a specific time course. The three VCVs [pa], [ta], and [ka], produced by one female speaker, were analyzed and reconstructed with the developed model. They were shown to be reasonable approximations of the original VCVs, as assessed qualitatively by visual inspection and quantitatively by calculating rms error and correlation coefficients. This establishes a method for future modeling of other speech material.

Original languageEnglish (US)
Pages (from-to)825-836
Number of pages12
JournalJournal of the Acoustical Society of America
Volume126
Issue number2
StatePublished - Aug 2009

Fingerprint

vowels
X-Rays
microbeams
coefficients
correlation coefficients
inspection
Vocal Tract
Consonant
perturbation
approximation
Superposition
x rays

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Vowel and consonant contributions to vocal tract shape. / Story, Brad H.

In: Journal of the Acoustical Society of America, Vol. 126, No. 2, 08.2009, p. 825-836.

Research output: Contribution to journalArticle

@article{46e6fb1d7d4e42f8a680482b2fdf2049,
title = "Vowel and consonant contributions to vocal tract shape.",
abstract = "The purpose of this study was to develop a method by which a vowel-consonant-vowel (VCV) utterance based on x-ray microbeam articulatory data could be separated into a vowel-to-vowel transition and a consonant superposition function. The result is a model that represents a vowel sequence as a time-dependent perturbation of the neutral vocal tract shape governed by coefficients of canonical deformation patterns. Consonants were modeled as superposition functions that can force specific portions of the vocal tract shape to be constricted or expanded, over a specific time course. The three VCVs [pa], [ta], and [ka], produced by one female speaker, were analyzed and reconstructed with the developed model. They were shown to be reasonable approximations of the original VCVs, as assessed qualitatively by visual inspection and quantitatively by calculating rms error and correlation coefficients. This establishes a method for future modeling of other speech material.",
author = "Story, {Brad H}",
year = "2009",
month = "8",
language = "English (US)",
volume = "126",
pages = "825--836",
journal = "Journal of the Acoustical Society of America",
issn = "0001-4966",
publisher = "Acoustical Society of America",
number = "2",

}

TY - JOUR

T1 - Vowel and consonant contributions to vocal tract shape.

AU - Story, Brad H

PY - 2009/8

Y1 - 2009/8

N2 - The purpose of this study was to develop a method by which a vowel-consonant-vowel (VCV) utterance based on x-ray microbeam articulatory data could be separated into a vowel-to-vowel transition and a consonant superposition function. The result is a model that represents a vowel sequence as a time-dependent perturbation of the neutral vocal tract shape governed by coefficients of canonical deformation patterns. Consonants were modeled as superposition functions that can force specific portions of the vocal tract shape to be constricted or expanded, over a specific time course. The three VCVs [pa], [ta], and [ka], produced by one female speaker, were analyzed and reconstructed with the developed model. They were shown to be reasonable approximations of the original VCVs, as assessed qualitatively by visual inspection and quantitatively by calculating rms error and correlation coefficients. This establishes a method for future modeling of other speech material.

AB - The purpose of this study was to develop a method by which a vowel-consonant-vowel (VCV) utterance based on x-ray microbeam articulatory data could be separated into a vowel-to-vowel transition and a consonant superposition function. The result is a model that represents a vowel sequence as a time-dependent perturbation of the neutral vocal tract shape governed by coefficients of canonical deformation patterns. Consonants were modeled as superposition functions that can force specific portions of the vocal tract shape to be constricted or expanded, over a specific time course. The three VCVs [pa], [ta], and [ka], produced by one female speaker, were analyzed and reconstructed with the developed model. They were shown to be reasonable approximations of the original VCVs, as assessed qualitatively by visual inspection and quantitatively by calculating rms error and correlation coefficients. This establishes a method for future modeling of other speech material.

UR - http://www.scopus.com/inward/record.url?scp=70349785878&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70349785878&partnerID=8YFLogxK

M3 - Article

C2 - 19640047

AN - SCOPUS:70349785878

VL - 126

SP - 825

EP - 836

JO - Journal of the Acoustical Society of America

JF - Journal of the Acoustical Society of America

SN - 0001-4966

IS - 2

ER -