Dysarthric speaker identification with different degrees of dysarthria severity using deep belief networks
Dysarthria is a degenerative disorder of the central nervous system that affects the control of articulation and pitch; therefore, it affects the uniqueness of sound produced by the speaker. Hence, dysarthric speaker recognition is a challenging task. In this paper, a feature‐extraction method based...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Electronics and Telecommunications Research Institute (ETRI)
2018-07-01
|
Series: | ETRI Journal |
Subjects: | |
Online Access: | https://doi.org/10.4218/etrij.2017-0260 |
id |
doaj-b01993f723b840e1bf4481bf4c3b62d0 |
---|---|
record_format |
Article |
spelling |
doaj-b01993f723b840e1bf4481bf4c3b62d02020-11-25T02:24:37ZengElectronics and Telecommunications Research Institute (ETRI)ETRI Journal1225-64632233-73262018-07-0140564365210.4218/etrij.2017-026010.4218/etrij.2017-0260Dysarthric speaker identification with different degrees of dysarthria severity using deep belief networksAref FarhadipourHadi VeisiMohammad AsgariMohammad Ali KeyvanradDysarthria is a degenerative disorder of the central nervous system that affects the control of articulation and pitch; therefore, it affects the uniqueness of sound produced by the speaker. Hence, dysarthric speaker recognition is a challenging task. In this paper, a feature‐extraction method based on deep belief networks is presented for the task of identifying a speaker suffering from dysarthria. The effectiveness of the proposed method is demonstrated and compared with well‐known Mel‐frequency cepstral coefficient features. For classification purposes, the use of a multi‐layer perceptron neural network is proposed with two structures. Our evaluations using the universal access speech database produced promising results and outperformed other baseline methods. In addition, speaker identification under both text‐dependent and text‐independent conditions are explored. The highest accuracy achieved using the proposed system is 97.3%.https://doi.org/10.4218/etrij.2017-0260deep belief networkdeep neural networkdysarthriaMFCCspeaker identification |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Aref Farhadipour Hadi Veisi Mohammad Asgari Mohammad Ali Keyvanrad |
spellingShingle |
Aref Farhadipour Hadi Veisi Mohammad Asgari Mohammad Ali Keyvanrad Dysarthric speaker identification with different degrees of dysarthria severity using deep belief networks ETRI Journal deep belief network deep neural network dysarthria MFCC speaker identification |
author_facet |
Aref Farhadipour Hadi Veisi Mohammad Asgari Mohammad Ali Keyvanrad |
author_sort |
Aref Farhadipour |
title |
Dysarthric speaker identification with different degrees of dysarthria severity using deep belief networks |
title_short |
Dysarthric speaker identification with different degrees of dysarthria severity using deep belief networks |
title_full |
Dysarthric speaker identification with different degrees of dysarthria severity using deep belief networks |
title_fullStr |
Dysarthric speaker identification with different degrees of dysarthria severity using deep belief networks |
title_full_unstemmed |
Dysarthric speaker identification with different degrees of dysarthria severity using deep belief networks |
title_sort |
dysarthric speaker identification with different degrees of dysarthria severity using deep belief networks |
publisher |
Electronics and Telecommunications Research Institute (ETRI) |
series |
ETRI Journal |
issn |
1225-6463 2233-7326 |
publishDate |
2018-07-01 |
description |
Dysarthria is a degenerative disorder of the central nervous system that affects the control of articulation and pitch; therefore, it affects the uniqueness of sound produced by the speaker. Hence, dysarthric speaker recognition is a challenging task. In this paper, a feature‐extraction method based on deep belief networks is presented for the task of identifying a speaker suffering from dysarthria. The effectiveness of the proposed method is demonstrated and compared with well‐known Mel‐frequency cepstral coefficient features. For classification purposes, the use of a multi‐layer perceptron neural network is proposed with two structures. Our evaluations using the universal access speech database produced promising results and outperformed other baseline methods. In addition, speaker identification under both text‐dependent and text‐independent conditions are explored. The highest accuracy achieved using the proposed system is 97.3%. |
topic |
deep belief network deep neural network dysarthria MFCC speaker identification |
url |
https://doi.org/10.4218/etrij.2017-0260 |
work_keys_str_mv |
AT areffarhadipour dysarthricspeakeridentificationwithdifferentdegreesofdysarthriaseverityusingdeepbeliefnetworks AT hadiveisi dysarthricspeakeridentificationwithdifferentdegreesofdysarthriaseverityusingdeepbeliefnetworks AT mohammadasgari dysarthricspeakeridentificationwithdifferentdegreesofdysarthriaseverityusingdeepbeliefnetworks AT mohammadalikeyvanrad dysarthricspeakeridentificationwithdifferentdegreesofdysarthriaseverityusingdeepbeliefnetworks |
_version_ |
1724854588932096000 |