Publication

Inducing a measure of phonetic similarity from pronunciation variation

Wieling, M., Margaretha, E. & Nerbonne, J., Mar-2012, In : Journal of Phonetics. 40, 2, p. 307-314 8 p.

Research output: Contribution to journalArticleAcademicpeer-review

APA

Wieling, M., Margaretha, E., & Nerbonne, J. (2012). Inducing a measure of phonetic similarity from pronunciation variation. Journal of Phonetics, 40(2), 307-314. https://doi.org/10.1016/j.wocn.2011.12.004

Author

Wieling, Martijn ; Margaretha, Eliza ; Nerbonne, John. / Inducing a measure of phonetic similarity from pronunciation variation. In: Journal of Phonetics. 2012 ; Vol. 40, No. 2. pp. 307-314.

Harvard

Wieling, M, Margaretha, E & Nerbonne, J 2012, 'Inducing a measure of phonetic similarity from pronunciation variation', Journal of Phonetics, vol. 40, no. 2, pp. 307-314. https://doi.org/10.1016/j.wocn.2011.12.004

Standard

Inducing a measure of phonetic similarity from pronunciation variation. / Wieling, Martijn; Margaretha, Eliza; Nerbonne, John.

In: Journal of Phonetics, Vol. 40, No. 2, 03.2012, p. 307-314.

Research output: Contribution to journalArticleAcademicpeer-review

Vancouver

Wieling M, Margaretha E, Nerbonne J. Inducing a measure of phonetic similarity from pronunciation variation. Journal of Phonetics. 2012 Mar;40(2):307-314. https://doi.org/10.1016/j.wocn.2011.12.004


BibTeX

@article{85cbd34717f94769b4a9bb6d1857546f,
title = "Inducing a measure of phonetic similarity from pronunciation variation",
abstract = "Structuralists famously observed that language is {"}un systeme oil tout se tient{"} (Meillet, 1903, p.407), insisting that the system of relations of linguistic units was more important than their concrete content. This study attempts to derive content from relations, in particular phonetic (acoustic) content from the distribution of alternative pronunciations used in different geographical varieties. It proceeds from data documenting language variation, examining six dialect atlases each containing the phonetic transcriptions of the same sets of words at hundreds of different sites. We obtain the sound segment correspondences via an alignment procedure, and then apply an information-theoretic measure, pointwise mutual information, assigning smaller segment distances to sound segment pairs which correspond relatively frequently. We iterate alignment and information-theoretic distance assignment until both remain stable, and we evaluate the quality of the resulting phonetic distances by comparing them to acoustic vowel distances. Wieling, Margaretha, and Nerbonne (2011) evaluated this method on the basis of Dutch and German dialect data, and here we provide more general support for the method by applying it to several other dialect datasets (i.e. Gabon Bantu, U.S. English, Tuscan and Bulgarian). We find relatively strong significant correlations between the induced phonetic distances and the acoustic distances, illustrating the usefulness of the method in deriving valid phonetic distances from distributions of dialectal variation. (C) 2011 Elsevier Ltd. All rights reserved.",
keywords = "FOREIGN ACCENT, ENGLISH, VOWELS, CLASSIFICATION, SPEAKERS",
author = "Martijn Wieling and Eliza Margaretha and John Nerbonne",
year = "2012",
month = mar,
doi = "10.1016/j.wocn.2011.12.004",
language = "English",
volume = "40",
pages = "307--314",
journal = "Journal of Phonetics",
issn = "0095-4470",
publisher = "ACADEMIC PRESS LTD- ELSEVIER SCIENCE LTD",
number = "2",

}

RIS

TY - JOUR

T1 - Inducing a measure of phonetic similarity from pronunciation variation

AU - Wieling, Martijn

AU - Margaretha, Eliza

AU - Nerbonne, John

PY - 2012/3

Y1 - 2012/3

N2 - Structuralists famously observed that language is "un systeme oil tout se tient" (Meillet, 1903, p.407), insisting that the system of relations of linguistic units was more important than their concrete content. This study attempts to derive content from relations, in particular phonetic (acoustic) content from the distribution of alternative pronunciations used in different geographical varieties. It proceeds from data documenting language variation, examining six dialect atlases each containing the phonetic transcriptions of the same sets of words at hundreds of different sites. We obtain the sound segment correspondences via an alignment procedure, and then apply an information-theoretic measure, pointwise mutual information, assigning smaller segment distances to sound segment pairs which correspond relatively frequently. We iterate alignment and information-theoretic distance assignment until both remain stable, and we evaluate the quality of the resulting phonetic distances by comparing them to acoustic vowel distances. Wieling, Margaretha, and Nerbonne (2011) evaluated this method on the basis of Dutch and German dialect data, and here we provide more general support for the method by applying it to several other dialect datasets (i.e. Gabon Bantu, U.S. English, Tuscan and Bulgarian). We find relatively strong significant correlations between the induced phonetic distances and the acoustic distances, illustrating the usefulness of the method in deriving valid phonetic distances from distributions of dialectal variation. (C) 2011 Elsevier Ltd. All rights reserved.

AB - Structuralists famously observed that language is "un systeme oil tout se tient" (Meillet, 1903, p.407), insisting that the system of relations of linguistic units was more important than their concrete content. This study attempts to derive content from relations, in particular phonetic (acoustic) content from the distribution of alternative pronunciations used in different geographical varieties. It proceeds from data documenting language variation, examining six dialect atlases each containing the phonetic transcriptions of the same sets of words at hundreds of different sites. We obtain the sound segment correspondences via an alignment procedure, and then apply an information-theoretic measure, pointwise mutual information, assigning smaller segment distances to sound segment pairs which correspond relatively frequently. We iterate alignment and information-theoretic distance assignment until both remain stable, and we evaluate the quality of the resulting phonetic distances by comparing them to acoustic vowel distances. Wieling, Margaretha, and Nerbonne (2011) evaluated this method on the basis of Dutch and German dialect data, and here we provide more general support for the method by applying it to several other dialect datasets (i.e. Gabon Bantu, U.S. English, Tuscan and Bulgarian). We find relatively strong significant correlations between the induced phonetic distances and the acoustic distances, illustrating the usefulness of the method in deriving valid phonetic distances from distributions of dialectal variation. (C) 2011 Elsevier Ltd. All rights reserved.

KW - FOREIGN ACCENT

KW - ENGLISH

KW - VOWELS

KW - CLASSIFICATION

KW - SPEAKERS

U2 - 10.1016/j.wocn.2011.12.004

DO - 10.1016/j.wocn.2011.12.004

M3 - Article

VL - 40

SP - 307

EP - 314

JO - Journal of Phonetics

JF - Journal of Phonetics

SN - 0095-4470

IS - 2

ER -

ID: 5519407