Application of neural networks to articulatory motion estimation

Tetsunori Kobayashi, Masayuki Yagyu, Katsuhiko Shirai

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

5 Citations (Scopus)

Abstract

The authors discuss an application of neural networks (NNs) to the problem of estimating the motion of articulatory organs from speech waves. A four-layer feedforward network was successfully applied to the articulatory parameter estimation problem. The evaluation test was performed using the vowel data in 5200 tokens in the ATR word database. Results show that the difference in estimated articulatory parameter values between the conventional model matching method (MM) and NN is only 0.1, which is about 3% of the value range, on average. For a few data, large differences arise between MM and NN, but this is due to misestimation in MM rather than NN. The percentage of misestimates in NN is less than 50% of that for MM. As for calculation time, NN is 10 times faster than MM. Thus, a high-speed and stable articulatory parameter estimation technique can be realized using neural networks.

Original languageEnglish
Title of host publicationProceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing
Editors Anon
Place of PublicationPiscataway, NJ, United States
PublisherPubl by IEEE
Pages489-492
Number of pages4
Volume1
ISBN (Print)078030033
Publication statusPublished - 1991
Externally publishedYes
EventProceedings of the 1991 International Conference on Acoustics, Speech, and Signal Processing - ICASSP 91 - Toronto, Ont, Can
Duration: 1991 May 141991 May 17

Other

OtherProceedings of the 1991 International Conference on Acoustics, Speech, and Signal Processing - ICASSP 91
CityToronto, Ont, Can
Period91/5/1491/5/17

Fingerprint

Motion estimation
Neural networks
Parameter estimation
vowels
organs
estimating
high speed
evaluation

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Acoustics and Ultrasonics

Cite this

Kobayashi, T., Yagyu, M., & Shirai, K. (1991). Application of neural networks to articulatory motion estimation. In Anon (Ed.), Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing (Vol. 1, pp. 489-492). Piscataway, NJ, United States: Publ by IEEE.

Application of neural networks to articulatory motion estimation. / Kobayashi, Tetsunori; Yagyu, Masayuki; Shirai, Katsuhiko.

Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing. ed. / Anon. Vol. 1 Piscataway, NJ, United States : Publ by IEEE, 1991. p. 489-492.

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

Kobayashi, T, Yagyu, M & Shirai, K 1991, Application of neural networks to articulatory motion estimation. in Anon (ed.), Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing. vol. 1, Publ by IEEE, Piscataway, NJ, United States, pp. 489-492, Proceedings of the 1991 International Conference on Acoustics, Speech, and Signal Processing - ICASSP 91, Toronto, Ont, Can, 91/5/14.
Kobayashi T, Yagyu M, Shirai K. Application of neural networks to articulatory motion estimation. In Anon, editor, Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing. Vol. 1. Piscataway, NJ, United States: Publ by IEEE. 1991. p. 489-492
Kobayashi, Tetsunori ; Yagyu, Masayuki ; Shirai, Katsuhiko. / Application of neural networks to articulatory motion estimation. Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing. editor / Anon. Vol. 1 Piscataway, NJ, United States : Publ by IEEE, 1991. pp. 489-492
@inproceedings{1769f859b64c4e71822898b29fa369d3,
title = "Application of neural networks to articulatory motion estimation",
abstract = "The authors discuss an application of neural networks (NNs) to the problem of estimating the motion of articulatory organs from speech waves. A four-layer feedforward network was successfully applied to the articulatory parameter estimation problem. The evaluation test was performed using the vowel data in 5200 tokens in the ATR word database. Results show that the difference in estimated articulatory parameter values between the conventional model matching method (MM) and NN is only 0.1, which is about 3{\%} of the value range, on average. For a few data, large differences arise between MM and NN, but this is due to misestimation in MM rather than NN. The percentage of misestimates in NN is less than 50{\%} of that for MM. As for calculation time, NN is 10 times faster than MM. Thus, a high-speed and stable articulatory parameter estimation technique can be realized using neural networks.",
author = "Tetsunori Kobayashi and Masayuki Yagyu and Katsuhiko Shirai",
year = "1991",
language = "English",
isbn = "078030033",
volume = "1",
pages = "489--492",
editor = "Anon",
booktitle = "Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing",
publisher = "Publ by IEEE",

}

TY - GEN

T1 - Application of neural networks to articulatory motion estimation

AU - Kobayashi, Tetsunori

AU - Yagyu, Masayuki

AU - Shirai, Katsuhiko

PY - 1991

Y1 - 1991

N2 - The authors discuss an application of neural networks (NNs) to the problem of estimating the motion of articulatory organs from speech waves. A four-layer feedforward network was successfully applied to the articulatory parameter estimation problem. The evaluation test was performed using the vowel data in 5200 tokens in the ATR word database. Results show that the difference in estimated articulatory parameter values between the conventional model matching method (MM) and NN is only 0.1, which is about 3% of the value range, on average. For a few data, large differences arise between MM and NN, but this is due to misestimation in MM rather than NN. The percentage of misestimates in NN is less than 50% of that for MM. As for calculation time, NN is 10 times faster than MM. Thus, a high-speed and stable articulatory parameter estimation technique can be realized using neural networks.

AB - The authors discuss an application of neural networks (NNs) to the problem of estimating the motion of articulatory organs from speech waves. A four-layer feedforward network was successfully applied to the articulatory parameter estimation problem. The evaluation test was performed using the vowel data in 5200 tokens in the ATR word database. Results show that the difference in estimated articulatory parameter values between the conventional model matching method (MM) and NN is only 0.1, which is about 3% of the value range, on average. For a few data, large differences arise between MM and NN, but this is due to misestimation in MM rather than NN. The percentage of misestimates in NN is less than 50% of that for MM. As for calculation time, NN is 10 times faster than MM. Thus, a high-speed and stable articulatory parameter estimation technique can be realized using neural networks.

UR - http://www.scopus.com/inward/record.url?scp=0026396340&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026396340&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0026396340

SN - 078030033

VL - 1

SP - 489

EP - 492

BT - Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing

A2 - Anon, null

PB - Publ by IEEE

CY - Piscataway, NJ, United States

ER -