Using a genetic algorithm to estimate parameters of a coarticulation model

Brian O. Bush, John Paul Hosom, Alexander Kain, Akiko Amano-Kusumoto

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

3 Citations (Scopus)

Abstract

We present a real-coded genetic algorithm that efficiently estimates parameters of a formant trajectory model. The genetic algorithm uses roulette-wheel selection and elitism to minimize the root mean square error between the observed formant trajectory and the model trajectory. Parameters, including vowel and consonant target values and coarticulation parameters, are estimated for a corpus of English clear and conversational CVC words. Results show consistent consonant formant targets, even when those consonants do not themselves have formant structure. We also present findings of a relationship between a coarticulation parameter and the consonant identity.

Original languageEnglish (US)
Title of host publicationProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Pages2677-2680
Number of pages4
StatePublished - 2011
Event12th Annual Conference of the International Speech Communication Association, INTERSPEECH 2011 - Florence, Italy
Duration: Aug 27 2011Aug 31 2011

Other

Other12th Annual Conference of the International Speech Communication Association, INTERSPEECH 2011
CountryItaly
CityFlorence
Period8/27/118/31/11

Fingerprint

Genetic algorithms
Trajectories
Genetic Algorithm
Trajectory
Estimate
Roulette
Real-coded Genetic Algorithm
Mean square error
Target
Wheels
Model
Wheel
Roots
Minimise
Formants
Coarticulation
Consonant

Keywords

  • Coarticulation
  • Formant targets
  • Genetic algorithms

ASJC Scopus subject areas

  • Language and Linguistics
  • Human-Computer Interaction
  • Signal Processing
  • Software
  • Modeling and Simulation

Cite this

Bush, B. O., Hosom, J. P., Kain, A., & Amano-Kusumoto, A. (2011). Using a genetic algorithm to estimate parameters of a coarticulation model. In Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH (pp. 2677-2680)

Using a genetic algorithm to estimate parameters of a coarticulation model. / Bush, Brian O.; Hosom, John Paul; Kain, Alexander; Amano-Kusumoto, Akiko.

Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH. 2011. p. 2677-2680.

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

Bush, BO, Hosom, JP, Kain, A & Amano-Kusumoto, A 2011, Using a genetic algorithm to estimate parameters of a coarticulation model. in Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH. pp. 2677-2680, 12th Annual Conference of the International Speech Communication Association, INTERSPEECH 2011, Florence, Italy, 8/27/11.
Bush BO, Hosom JP, Kain A, Amano-Kusumoto A. Using a genetic algorithm to estimate parameters of a coarticulation model. In Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH. 2011. p. 2677-2680
Bush, Brian O. ; Hosom, John Paul ; Kain, Alexander ; Amano-Kusumoto, Akiko. / Using a genetic algorithm to estimate parameters of a coarticulation model. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH. 2011. pp. 2677-2680
@inproceedings{adb52c824c894110a253a547d37dbd5b,
title = "Using a genetic algorithm to estimate parameters of a coarticulation model",
abstract = "We present a real-coded genetic algorithm that efficiently estimates parameters of a formant trajectory model. The genetic algorithm uses roulette-wheel selection and elitism to minimize the root mean square error between the observed formant trajectory and the model trajectory. Parameters, including vowel and consonant target values and coarticulation parameters, are estimated for a corpus of English clear and conversational CVC words. Results show consistent consonant formant targets, even when those consonants do not themselves have formant structure. We also present findings of a relationship between a coarticulation parameter and the consonant identity.",
keywords = "Coarticulation, Formant targets, Genetic algorithms",
author = "Bush, {Brian O.} and Hosom, {John Paul} and Alexander Kain and Akiko Amano-Kusumoto",
year = "2011",
language = "English (US)",
pages = "2677--2680",
booktitle = "Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH",

}

TY - GEN

T1 - Using a genetic algorithm to estimate parameters of a coarticulation model

AU - Bush, Brian O.

AU - Hosom, John Paul

AU - Kain, Alexander

AU - Amano-Kusumoto, Akiko

PY - 2011

Y1 - 2011

N2 - We present a real-coded genetic algorithm that efficiently estimates parameters of a formant trajectory model. The genetic algorithm uses roulette-wheel selection and elitism to minimize the root mean square error between the observed formant trajectory and the model trajectory. Parameters, including vowel and consonant target values and coarticulation parameters, are estimated for a corpus of English clear and conversational CVC words. Results show consistent consonant formant targets, even when those consonants do not themselves have formant structure. We also present findings of a relationship between a coarticulation parameter and the consonant identity.

AB - We present a real-coded genetic algorithm that efficiently estimates parameters of a formant trajectory model. The genetic algorithm uses roulette-wheel selection and elitism to minimize the root mean square error between the observed formant trajectory and the model trajectory. Parameters, including vowel and consonant target values and coarticulation parameters, are estimated for a corpus of English clear and conversational CVC words. Results show consistent consonant formant targets, even when those consonants do not themselves have formant structure. We also present findings of a relationship between a coarticulation parameter and the consonant identity.

KW - Coarticulation

KW - Formant targets

KW - Genetic algorithms

UR - http://www.scopus.com/inward/record.url?scp=84865726819&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865726819&partnerID=8YFLogxK

M3 - Conference contribution

SP - 2677

EP - 2680

BT - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

ER -