Machine Learning Based Automated Speech Dialog Analysis Of Autistic Children

Anjana  Wijesinghe; Pradeepa  Samarasinghe; Sudarshi  Seneviratne; Y Pratheepan; Koliya  Pulasinghe

doi:10.1109/KSE.2019.8919266

Machine Learning Based Automated Speech Dialog Analysis Of Autistic Children

Anjana Wijesinghe, Pradeepa Samarasinghe, Sudarshi Seneviratne, Y Pratheepan, Koliya Pulasinghe

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

11 Citations (Scopus)

491 Downloads (Pure)

Abstract

Children with autism spectrum disorder (ASD) have altered behaviors in communication, social interaction, and activity, out of which communication has been the most prominent disorder among many. Despite the recent technological advances, limited attention has been given to screening and diagnosing ASD by identifying the speech deficiencies (SD) of autistic children
at early stages. This research focuses on bridging the gap in ASD screening by developing an automated system to distinguish autistic traits through speech analysis. Data was collected from 40 participants for the initial analysis and recordings were obtained from 17 participants. We considered a three-stage processing system; first stage utilizes thresholding for silence detection and
Vocal Activity Detection for vocal isolation, second stage adopts machine learning technique neural network with frequency domain representations in developing a reliant utterance classifier for the isolated vocals and stage three also adopts machine learning technique neural network in recognizing autistic traits in speech patterns of the classified utterances. The results are promising in identifying SD of autistic children with the utterance classifier having 78% accuracy and pattern recognition 72% accuracy.

Original language	English
Title of host publication	Proceedings of 2019 11th International Conference on Knowledge and Systems Engineering, KSE 2019
Editors	Josiane Mothe, Le Hoang Son, Nguyen Tran Quoc Vinh
Place of Publication	Da Nang, Vietnam, Vietnam
Publisher	IEEE
Pages	1-5
Number of pages	5
ISBN (Electronic)	978-1-7281-3003-3
ISBN (Print)	978-1-7281-3004-0
DOIs	https://doi.org/10.1109/KSE.2019.8919266
Publication status	Published (in print/issue) - 24 Oct 2019

Keywords

Audio Classification
Autism Screening
Neural Networks
Speech Deficiencies
Speech Pattern Recognition

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/KSE.2019.8919266

KSE__19Author Accepted Manuscript, 347 KBLicence: Other
08919266Final published version, 1.05 MB

Cite this

Wijesinghe, A., Samarasinghe, P., Seneviratne, S., Pratheepan, Y., & Pulasinghe, K. (2019). Machine Learning Based Automated Speech Dialog Analysis Of Autistic Children. In J. Mothe, L. H. Son, & N. T. Q. Vinh (Eds.), Proceedings of 2019 11th International Conference on Knowledge and Systems Engineering, KSE 2019 (pp. 1-5). Article 8919266 IEEE. https://doi.org/10.1109/KSE.2019.8919266

@inproceedings{a9a74ac8ac1b429085d6d2e3c2232c98,

title = "Machine Learning Based Automated Speech Dialog Analysis Of Autistic Children",

abstract = "Children with autism spectrum disorder (ASD) have altered behaviors in communication, social interaction, and activity, out of which communication has been the most prominent disorder among many. Despite the recent technological advances, limited attention has been given to screening and diagnosing ASD by identifying the speech deficiencies (SD) of autistic childrenat early stages. This research focuses on bridging the gap in ASD screening by developing an automated system to distinguish autistic traits through speech analysis. Data was collected from 40 participants for the initial analysis and recordings were obtained from 17 participants. We considered a three-stage processing system; first stage utilizes thresholding for silence detection andVocal Activity Detection for vocal isolation, second stage adopts machine learning technique neural network with frequency domain representations in developing a reliant utterance classifier for the isolated vocals and stage three also adopts machine learning technique neural network in recognizing autistic traits in speech patterns of the classified utterances. The results are promising in identifying SD of autistic children with the utterance classifier having 78% accuracy and pattern recognition 72% accuracy.",

keywords = "Audio Classification, Autism Screening, Neural Networks, Speech Deficiencies, Speech Pattern Recognition",

author = "Anjana Wijesinghe and Pradeepa Samarasinghe and Sudarshi Seneviratne and Y Pratheepan and Koliya Pulasinghe",

year = "2019",

month = oct,

day = "24",

doi = "10.1109/KSE.2019.8919266",

language = "English",

isbn = "978-1-7281-3004-0",

pages = "1--5",

editor = "Josiane Mothe and Son, {Le Hoang} and Vinh, {Nguyen Tran Quoc}",

booktitle = "Proceedings of 2019 11th International Conference on Knowledge and Systems Engineering, KSE 2019",

publisher = "IEEE",

address = "United States",

}

Wijesinghe, A, Samarasinghe, P, Seneviratne, S, Pratheepan, Y & Pulasinghe, K 2019, Machine Learning Based Automated Speech Dialog Analysis Of Autistic Children. in J Mothe, LH Son & NTQ Vinh (eds), Proceedings of 2019 11th International Conference on Knowledge and Systems Engineering, KSE 2019., 8919266, IEEE, Da Nang, Vietnam, Vietnam, pp. 1-5. https://doi.org/10.1109/KSE.2019.8919266

Machine Learning Based Automated Speech Dialog Analysis Of Autistic Children. / Wijesinghe, Anjana ; Samarasinghe, Pradeepa; Seneviratne, Sudarshi et al.
Proceedings of 2019 11th International Conference on Knowledge and Systems Engineering, KSE 2019. ed. / Josiane Mothe; Le Hoang Son; Nguyen Tran Quoc Vinh. Da Nang, Vietnam, Vietnam: IEEE, 2019. p. 1-5 8919266.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Machine Learning Based Automated Speech Dialog Analysis Of Autistic Children

AU - Wijesinghe, Anjana

AU - Samarasinghe, Pradeepa

AU - Seneviratne, Sudarshi

AU - Pratheepan, Y

AU - Pulasinghe, Koliya

PY - 2019/10/24

Y1 - 2019/10/24

N2 - Children with autism spectrum disorder (ASD) have altered behaviors in communication, social interaction, and activity, out of which communication has been the most prominent disorder among many. Despite the recent technological advances, limited attention has been given to screening and diagnosing ASD by identifying the speech deficiencies (SD) of autistic childrenat early stages. This research focuses on bridging the gap in ASD screening by developing an automated system to distinguish autistic traits through speech analysis. Data was collected from 40 participants for the initial analysis and recordings were obtained from 17 participants. We considered a three-stage processing system; first stage utilizes thresholding for silence detection andVocal Activity Detection for vocal isolation, second stage adopts machine learning technique neural network with frequency domain representations in developing a reliant utterance classifier for the isolated vocals and stage three also adopts machine learning technique neural network in recognizing autistic traits in speech patterns of the classified utterances. The results are promising in identifying SD of autistic children with the utterance classifier having 78% accuracy and pattern recognition 72% accuracy.

AB - Children with autism spectrum disorder (ASD) have altered behaviors in communication, social interaction, and activity, out of which communication has been the most prominent disorder among many. Despite the recent technological advances, limited attention has been given to screening and diagnosing ASD by identifying the speech deficiencies (SD) of autistic childrenat early stages. This research focuses on bridging the gap in ASD screening by developing an automated system to distinguish autistic traits through speech analysis. Data was collected from 40 participants for the initial analysis and recordings were obtained from 17 participants. We considered a three-stage processing system; first stage utilizes thresholding for silence detection andVocal Activity Detection for vocal isolation, second stage adopts machine learning technique neural network with frequency domain representations in developing a reliant utterance classifier for the isolated vocals and stage three also adopts machine learning technique neural network in recognizing autistic traits in speech patterns of the classified utterances. The results are promising in identifying SD of autistic children with the utterance classifier having 78% accuracy and pattern recognition 72% accuracy.

KW - Audio Classification

KW - Autism Screening

KW - Neural Networks

KW - Speech Deficiencies

KW - Speech Pattern Recognition

UR - http://www.scopus.com/inward/record.url?scp=85077069435&partnerID=8YFLogxK

U2 - 10.1109/KSE.2019.8919266

DO - 10.1109/KSE.2019.8919266

M3 - Conference contribution

SN - 978-1-7281-3004-0

SP - 1

EP - 5

BT - Proceedings of 2019 11th International Conference on Knowledge and Systems Engineering, KSE 2019

A2 - Mothe, Josiane

A2 - Son, Le Hoang

A2 - Vinh, Nguyen Tran Quoc

PB - IEEE

CY - Da Nang, Vietnam, Vietnam

ER -

Wijesinghe A, Samarasinghe P, Seneviratne S, Pratheepan Y, Pulasinghe K. Machine Learning Based Automated Speech Dialog Analysis Of Autistic Children. In Mothe J, Son LH, Vinh NTQ, editors, Proceedings of 2019 11th International Conference on Knowledge and Systems Engineering, KSE 2019. Da Nang, Vietnam, Vietnam: IEEE. 2019. p. 1-5. 8919266 doi: 10.1109/KSE.2019.8919266

Machine Learning Based Automated Speech Dialog Analysis Of Autistic Children

Abstract

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this