Universal access: Speech recognition for talkers with spastic dysarthria

Harsh Vardhan Sharma, Mark Hasegawa-Johnson

Research output: Contribution to journalConference article

Abstract

This paper describes the results of our experiments in small and medium vocabulary dysarthric speech recognition, using the database being recorded by our group under the Universal Access initiative. We develop and test speaker-dependent, word- and phone-level speech recognizers utilizing the hidden Markov Model architecture; the models are trained exclusively on dysarthric speech produced by individuals diagnosed with cerebral palsy. The experiments indicate that (a) different system configurations (being word vs. phone based, number of states per HMM, number of Gaussian components per state specific observation probability density etc.) give useful performance (in terms of recognition accuracy) for different speakers and different task-vocabularies, and (b) for very low intelligibility subjects, speech recognition outperforms human listeners on recognizing dysarthric speech.

Original languageEnglish (US)
Pages (from-to)1451-1454
Number of pages4
JournalProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
StatePublished - Nov 26 2009
Event10th Annual Conference of the International Speech Communication Association, INTERSPEECH 2009 - Brighton, United Kingdom
Duration: Sep 6 2009Sep 10 2009

Fingerprint

Dysarthria
Speech recognition
Vocabulary
Speech intelligibility
Hidden Markov models
Speech Intelligibility
Cerebral Palsy
Experiments
Observation
Databases

Keywords

  • Assistive technology
  • Augmentative communication
  • Cerebral palsy
  • Dysarthria
  • Human-computer interface
  • Speech recognition

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Signal Processing
  • Software
  • Sensory Systems

Cite this

@article{276ecc8167444123a5367da90104b479,
title = "Universal access: Speech recognition for talkers with spastic dysarthria",
abstract = "This paper describes the results of our experiments in small and medium vocabulary dysarthric speech recognition, using the database being recorded by our group under the Universal Access initiative. We develop and test speaker-dependent, word- and phone-level speech recognizers utilizing the hidden Markov Model architecture; the models are trained exclusively on dysarthric speech produced by individuals diagnosed with cerebral palsy. The experiments indicate that (a) different system configurations (being word vs. phone based, number of states per HMM, number of Gaussian components per state specific observation probability density etc.) give useful performance (in terms of recognition accuracy) for different speakers and different task-vocabularies, and (b) for very low intelligibility subjects, speech recognition outperforms human listeners on recognizing dysarthric speech.",
keywords = "Assistive technology, Augmentative communication, Cerebral palsy, Dysarthria, Human-computer interface, Speech recognition",
author = "Sharma, {Harsh Vardhan} and Mark Hasegawa-Johnson",
year = "2009",
month = "11",
day = "26",
language = "English (US)",
pages = "1451--1454",
journal = "Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH",
issn = "2308-457X",

}

TY - JOUR

T1 - Universal access

T2 - Speech recognition for talkers with spastic dysarthria

AU - Sharma, Harsh Vardhan

AU - Hasegawa-Johnson, Mark

PY - 2009/11/26

Y1 - 2009/11/26

N2 - This paper describes the results of our experiments in small and medium vocabulary dysarthric speech recognition, using the database being recorded by our group under the Universal Access initiative. We develop and test speaker-dependent, word- and phone-level speech recognizers utilizing the hidden Markov Model architecture; the models are trained exclusively on dysarthric speech produced by individuals diagnosed with cerebral palsy. The experiments indicate that (a) different system configurations (being word vs. phone based, number of states per HMM, number of Gaussian components per state specific observation probability density etc.) give useful performance (in terms of recognition accuracy) for different speakers and different task-vocabularies, and (b) for very low intelligibility subjects, speech recognition outperforms human listeners on recognizing dysarthric speech.

AB - This paper describes the results of our experiments in small and medium vocabulary dysarthric speech recognition, using the database being recorded by our group under the Universal Access initiative. We develop and test speaker-dependent, word- and phone-level speech recognizers utilizing the hidden Markov Model architecture; the models are trained exclusively on dysarthric speech produced by individuals diagnosed with cerebral palsy. The experiments indicate that (a) different system configurations (being word vs. phone based, number of states per HMM, number of Gaussian components per state specific observation probability density etc.) give useful performance (in terms of recognition accuracy) for different speakers and different task-vocabularies, and (b) for very low intelligibility subjects, speech recognition outperforms human listeners on recognizing dysarthric speech.

KW - Assistive technology

KW - Augmentative communication

KW - Cerebral palsy

KW - Dysarthria

KW - Human-computer interface

KW - Speech recognition

UR - http://www.scopus.com/inward/record.url?scp=70450161730&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70450161730&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:70450161730

SP - 1451

EP - 1454

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

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

SN - 2308-457X

ER -