Unvoiced speech recognition using EMG - Mime Speech Recognition

Hiroyuki Manabe, Akira Hiraiwa, Toshiaki Sugimura

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

45 Citations (Scopus)

Abstract

We propose unvoiced speech recognition, "Mime Speech Recognition". It recognizes speech by observing the muscles associated with speech. It is not based on voice signals but electromyography (EMG). It will realize unvoiced communication, which is a new communication style. Because voice signals are not used, it can be applied in noisy environments; it also supports people without vocal-cords and aphasics. In preliminary experiments, we try to recognize the 5 Japanese vowels. EMG signals from the 3 muscles that contribute greatly to the utterance of Japanese vowels are input to a neural network. The recognition accuracy is over 90% for the three subjects tested.

Original languageEnglish
Title of host publicationCHI'03 Extended Abstracts on Human Factors in Computing Systems, CHI EA'03
Pages794-795
Number of pages2
DOIs
Publication statusPublished - 2003 Dec 1
Externally publishedYes
EventConference on Human Factors in Computing Systems, CHI EA 2003 - Ft. Lauderdale, FL, United States
Duration: 2003 Apr 52003 Apr 10

Publication series

NameConference on Human Factors in Computing Systems - Proceedings

Conference

ConferenceConference on Human Factors in Computing Systems, CHI EA 2003
CountryUnited States
CityFt. Lauderdale, FL
Period03/4/503/4/10

Fingerprint

Electromyography
Speech recognition
Muscle
Communication
Neural networks
Experiments

Keywords

  • EMG
  • Neural network
  • Speech recognition
  • Vowel

ASJC Scopus subject areas

  • Software
  • Human-Computer Interaction
  • Computer Graphics and Computer-Aided Design

Cite this

Manabe, H., Hiraiwa, A., & Sugimura, T. (2003). Unvoiced speech recognition using EMG - Mime Speech Recognition. In CHI'03 Extended Abstracts on Human Factors in Computing Systems, CHI EA'03 (pp. 794-795). (Conference on Human Factors in Computing Systems - Proceedings). https://doi.org/10.1145/765891.765996

Unvoiced speech recognition using EMG - Mime Speech Recognition. / Manabe, Hiroyuki; Hiraiwa, Akira; Sugimura, Toshiaki.

CHI'03 Extended Abstracts on Human Factors in Computing Systems, CHI EA'03. 2003. p. 794-795 (Conference on Human Factors in Computing Systems - Proceedings).

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

Manabe, H, Hiraiwa, A & Sugimura, T 2003, Unvoiced speech recognition using EMG - Mime Speech Recognition. in CHI'03 Extended Abstracts on Human Factors in Computing Systems, CHI EA'03. Conference on Human Factors in Computing Systems - Proceedings, pp. 794-795, Conference on Human Factors in Computing Systems, CHI EA 2003, Ft. Lauderdale, FL, United States, 03/4/5. https://doi.org/10.1145/765891.765996
Manabe H, Hiraiwa A, Sugimura T. Unvoiced speech recognition using EMG - Mime Speech Recognition. In CHI'03 Extended Abstracts on Human Factors in Computing Systems, CHI EA'03. 2003. p. 794-795. (Conference on Human Factors in Computing Systems - Proceedings). https://doi.org/10.1145/765891.765996
Manabe, Hiroyuki ; Hiraiwa, Akira ; Sugimura, Toshiaki. / Unvoiced speech recognition using EMG - Mime Speech Recognition. CHI'03 Extended Abstracts on Human Factors in Computing Systems, CHI EA'03. 2003. pp. 794-795 (Conference on Human Factors in Computing Systems - Proceedings).
@inproceedings{6b54ec2dc48c4ea4bc22c6a3b4d77458,
title = "Unvoiced speech recognition using EMG - Mime Speech Recognition",
abstract = "We propose unvoiced speech recognition, {"}Mime Speech Recognition{"}. It recognizes speech by observing the muscles associated with speech. It is not based on voice signals but electromyography (EMG). It will realize unvoiced communication, which is a new communication style. Because voice signals are not used, it can be applied in noisy environments; it also supports people without vocal-cords and aphasics. In preliminary experiments, we try to recognize the 5 Japanese vowels. EMG signals from the 3 muscles that contribute greatly to the utterance of Japanese vowels are input to a neural network. The recognition accuracy is over 90{\%} for the three subjects tested.",
keywords = "EMG, Neural network, Speech recognition, Vowel",
author = "Hiroyuki Manabe and Akira Hiraiwa and Toshiaki Sugimura",
year = "2003",
month = "12",
day = "1",
doi = "10.1145/765891.765996",
language = "English",
isbn = "1581136374",
series = "Conference on Human Factors in Computing Systems - Proceedings",
pages = "794--795",
booktitle = "CHI'03 Extended Abstracts on Human Factors in Computing Systems, CHI EA'03",

}

TY - GEN

T1 - Unvoiced speech recognition using EMG - Mime Speech Recognition

AU - Manabe, Hiroyuki

AU - Hiraiwa, Akira

AU - Sugimura, Toshiaki

PY - 2003/12/1

Y1 - 2003/12/1

N2 - We propose unvoiced speech recognition, "Mime Speech Recognition". It recognizes speech by observing the muscles associated with speech. It is not based on voice signals but electromyography (EMG). It will realize unvoiced communication, which is a new communication style. Because voice signals are not used, it can be applied in noisy environments; it also supports people without vocal-cords and aphasics. In preliminary experiments, we try to recognize the 5 Japanese vowels. EMG signals from the 3 muscles that contribute greatly to the utterance of Japanese vowels are input to a neural network. The recognition accuracy is over 90% for the three subjects tested.

AB - We propose unvoiced speech recognition, "Mime Speech Recognition". It recognizes speech by observing the muscles associated with speech. It is not based on voice signals but electromyography (EMG). It will realize unvoiced communication, which is a new communication style. Because voice signals are not used, it can be applied in noisy environments; it also supports people without vocal-cords and aphasics. In preliminary experiments, we try to recognize the 5 Japanese vowels. EMG signals from the 3 muscles that contribute greatly to the utterance of Japanese vowels are input to a neural network. The recognition accuracy is over 90% for the three subjects tested.

KW - EMG

KW - Neural network

KW - Speech recognition

KW - Vowel

UR - http://www.scopus.com/inward/record.url?scp=84869004619&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84869004619&partnerID=8YFLogxK

U2 - 10.1145/765891.765996

DO - 10.1145/765891.765996

M3 - Conference contribution

SN - 1581136374

SN - 9781581136371

T3 - Conference on Human Factors in Computing Systems - Proceedings

SP - 794

EP - 795

BT - CHI'03 Extended Abstracts on Human Factors in Computing Systems, CHI EA'03

ER -