Abstract

This paper will investigate viability of a screening application that could be used to identify individuals with Dysarthria from among a larger population using sentence-level speech data. This task presents a number of challenged particularly if we aim to identify the disorder in the earlier stages before the more significant symptoms have begun to manifest themselves. A principal challenge in this task is acheiving robustness to the large number of confounding variables such as gender, age, accent, speaking style, and recording conditions. All of these variables will affect an individuals speech in a manner unrelated to the disorder, and identifying what information is relevant to the disorder amongst these confounding variables given the limited amount of data that is available in this regime presents a major engineering challenge. In this paper we will focus on achieving robustness to different types and levels of noise by employing a feature selection algorithm that attempts to minimize a non-parametric upper bound of the error in the noisy condition. This is a crucial problem to solve as the clean recording conditions used in data collection are typically a poor reflection of the type of data that will be encountered upon deployment.

Original languageEnglish (US)
Title of host publication2016 Digital Media Industry and Academic Forum, DMIAF 2016 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages135-138
Number of pages4
ISBN (Electronic)9781509010004
DOIs
StatePublished - Sep 22 2016
Event2016 Digital Media Industry and Academic Forum, DMIAF 2016 - Santorini, Greece
Duration: Jul 4 2016Jul 6 2016

Other

Other2016 Digital Media Industry and Academic Forum, DMIAF 2016
CountryGreece
CitySantorini
Period7/4/167/6/16

Fingerprint

Feature extraction
Screening

ASJC Scopus subject areas

  • Computer Science Applications
  • Media Technology

Cite this

Wisler, A., Berisha, V., Spanias, A., & Liss, J. (2016). Noise robust dysarthric speech classification using domain adaptation. In 2016 Digital Media Industry and Academic Forum, DMIAF 2016 - Proceedings (pp. 135-138). [7574918] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/DMIAF.2016.7574918

Noise robust dysarthric speech classification using domain adaptation. / Wisler, Alan; Berisha, Visar; Spanias, Andreas; Liss, Julie.

2016 Digital Media Industry and Academic Forum, DMIAF 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2016. p. 135-138 7574918.

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

Wisler, A, Berisha, V, Spanias, A & Liss, J 2016, Noise robust dysarthric speech classification using domain adaptation. in 2016 Digital Media Industry and Academic Forum, DMIAF 2016 - Proceedings., 7574918, Institute of Electrical and Electronics Engineers Inc., pp. 135-138, 2016 Digital Media Industry and Academic Forum, DMIAF 2016, Santorini, Greece, 7/4/16. https://doi.org/10.1109/DMIAF.2016.7574918
Wisler A, Berisha V, Spanias A, Liss J. Noise robust dysarthric speech classification using domain adaptation. In 2016 Digital Media Industry and Academic Forum, DMIAF 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2016. p. 135-138. 7574918 https://doi.org/10.1109/DMIAF.2016.7574918
Wisler, Alan ; Berisha, Visar ; Spanias, Andreas ; Liss, Julie. / Noise robust dysarthric speech classification using domain adaptation. 2016 Digital Media Industry and Academic Forum, DMIAF 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 135-138
@inproceedings{26808c8a1022429e8471bd813f5cc998,
title = "Noise robust dysarthric speech classification using domain adaptation",
abstract = "This paper will investigate viability of a screening application that could be used to identify individuals with Dysarthria from among a larger population using sentence-level speech data. This task presents a number of challenged particularly if we aim to identify the disorder in the earlier stages before the more significant symptoms have begun to manifest themselves. A principal challenge in this task is acheiving robustness to the large number of confounding variables such as gender, age, accent, speaking style, and recording conditions. All of these variables will affect an individuals speech in a manner unrelated to the disorder, and identifying what information is relevant to the disorder amongst these confounding variables given the limited amount of data that is available in this regime presents a major engineering challenge. In this paper we will focus on achieving robustness to different types and levels of noise by employing a feature selection algorithm that attempts to minimize a non-parametric upper bound of the error in the noisy condition. This is a crucial problem to solve as the clean recording conditions used in data collection are typically a poor reflection of the type of data that will be encountered upon deployment.",
author = "Alan Wisler and Visar Berisha and Andreas Spanias and Julie Liss",
year = "2016",
month = "9",
day = "22",
doi = "10.1109/DMIAF.2016.7574918",
language = "English (US)",
pages = "135--138",
booktitle = "2016 Digital Media Industry and Academic Forum, DMIAF 2016 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - Noise robust dysarthric speech classification using domain adaptation

AU - Wisler, Alan

AU - Berisha, Visar

AU - Spanias, Andreas

AU - Liss, Julie

PY - 2016/9/22

Y1 - 2016/9/22

N2 - This paper will investigate viability of a screening application that could be used to identify individuals with Dysarthria from among a larger population using sentence-level speech data. This task presents a number of challenged particularly if we aim to identify the disorder in the earlier stages before the more significant symptoms have begun to manifest themselves. A principal challenge in this task is acheiving robustness to the large number of confounding variables such as gender, age, accent, speaking style, and recording conditions. All of these variables will affect an individuals speech in a manner unrelated to the disorder, and identifying what information is relevant to the disorder amongst these confounding variables given the limited amount of data that is available in this regime presents a major engineering challenge. In this paper we will focus on achieving robustness to different types and levels of noise by employing a feature selection algorithm that attempts to minimize a non-parametric upper bound of the error in the noisy condition. This is a crucial problem to solve as the clean recording conditions used in data collection are typically a poor reflection of the type of data that will be encountered upon deployment.

AB - This paper will investigate viability of a screening application that could be used to identify individuals with Dysarthria from among a larger population using sentence-level speech data. This task presents a number of challenged particularly if we aim to identify the disorder in the earlier stages before the more significant symptoms have begun to manifest themselves. A principal challenge in this task is acheiving robustness to the large number of confounding variables such as gender, age, accent, speaking style, and recording conditions. All of these variables will affect an individuals speech in a manner unrelated to the disorder, and identifying what information is relevant to the disorder amongst these confounding variables given the limited amount of data that is available in this regime presents a major engineering challenge. In this paper we will focus on achieving robustness to different types and levels of noise by employing a feature selection algorithm that attempts to minimize a non-parametric upper bound of the error in the noisy condition. This is a crucial problem to solve as the clean recording conditions used in data collection are typically a poor reflection of the type of data that will be encountered upon deployment.

UR - http://www.scopus.com/inward/record.url?scp=84991781143&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84991781143&partnerID=8YFLogxK

U2 - 10.1109/DMIAF.2016.7574918

DO - 10.1109/DMIAF.2016.7574918

M3 - Conference contribution

SP - 135

EP - 138

BT - 2016 Digital Media Industry and Academic Forum, DMIAF 2016 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -