Combined Method for Informative Feature Selection for Speech Pathology Detection

The task of detecting vocal abnormalities is characterized by a small amount of available data for training, as a consequence of which classification systems that use low-dimensional data are the most relevant. We propose to use LASSO (least absolute shrinkage and selection operator) and BSS (backwa...

Full description

Saved in:
Bibliographic Details
Main Authors: D. S. Likhachov, M. I. Vashkevich, N. A. Petrovsky, E. S. Azarov
Format: Article
Language:Russian
Published: Educational institution «Belarusian State University of Informatics and Radioelectronics» 2023-08-01
Series:Doklady Belorusskogo gosudarstvennogo universiteta informatiki i radioèlektroniki
Subjects:
Online Access:https://doklady.bsuir.by/jour/article/view/3689
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849240537848086528
author D. S. Likhachov
M. I. Vashkevich
N. A. Petrovsky
E. S. Azarov
author_facet D. S. Likhachov
M. I. Vashkevich
N. A. Petrovsky
E. S. Azarov
author_sort D. S. Likhachov
collection DOAJ
description The task of detecting vocal abnormalities is characterized by a small amount of available data for training, as a consequence of which classification systems that use low-dimensional data are the most relevant. We propose to use LASSO (least absolute shrinkage and selection operator) and BSS (backward stepwise selection) methods together to select the most significant features for the detection of vocal pathologies, in particular amyotrophic lateral sclerosis. Features based on fine-frequency cepstral coefficients, traditionally used in speech signal processing, and features based on discrete estimation of the autoregressive spectrum envelope are used. Spectral features based on the autoregressive process envelope spectrum are extracted using the generative method, which involves calculating a discrete Fourier transform of the report sequence generated using the autoregressive model of the input voice signal. The sequence is generated by the autoregressive model so as to account for the periodic nature of the Fourier transform. This improves the accuracy of the spectrum estimation and reduces the spectral leakage effect. Using LASSO in conjunction with BSS allowed us to improve the classification efficiency using a smaller number of features as compared to using the LASSO method alone.
format Article
id doaj-art-adc86e638e1b4ae4b317c0a2c504a89c
institution Kabale University
issn 1729-7648
language Russian
publishDate 2023-08-01
publisher Educational institution «Belarusian State University of Informatics and Radioelectronics»
record_format Article
series Doklady Belorusskogo gosudarstvennogo universiteta informatiki i radioèlektroniki
spelling doaj-art-adc86e638e1b4ae4b317c0a2c504a89c2025-08-20T04:00:33ZrusEducational institution «Belarusian State University of Informatics and Radioelectronics»Doklady Belorusskogo gosudarstvennogo universiteta informatiki i radioèlektroniki1729-76482023-08-0121411011710.35596/1729-7648-2023-21-4-110-1171928Combined Method for Informative Feature Selection for Speech Pathology DetectionD. S. Likhachov0M. I. Vashkevich1N. A. Petrovsky2E. S. Azarov3Belarusian State University of Informatics and RadioelectronicsBelarusian State University of Informatics and RadioelectronicsBelarusian State University of Informatics and RadioelectronicsBelarusian State University of Informatics and RadioelectronicsThe task of detecting vocal abnormalities is characterized by a small amount of available data for training, as a consequence of which classification systems that use low-dimensional data are the most relevant. We propose to use LASSO (least absolute shrinkage and selection operator) and BSS (backward stepwise selection) methods together to select the most significant features for the detection of vocal pathologies, in particular amyotrophic lateral sclerosis. Features based on fine-frequency cepstral coefficients, traditionally used in speech signal processing, and features based on discrete estimation of the autoregressive spectrum envelope are used. Spectral features based on the autoregressive process envelope spectrum are extracted using the generative method, which involves calculating a discrete Fourier transform of the report sequence generated using the autoregressive model of the input voice signal. The sequence is generated by the autoregressive model so as to account for the periodic nature of the Fourier transform. This improves the accuracy of the spectrum estimation and reduces the spectral leakage effect. Using LASSO in conjunction with BSS allowed us to improve the classification efficiency using a smaller number of features as compared to using the LASSO method alone.https://doklady.bsuir.by/jour/article/view/3689voice analysisgenerative methodautoregressionmachine learningspectral featuresclassification
spellingShingle D. S. Likhachov
M. I. Vashkevich
N. A. Petrovsky
E. S. Azarov
Combined Method for Informative Feature Selection for Speech Pathology Detection
Doklady Belorusskogo gosudarstvennogo universiteta informatiki i radioèlektroniki
voice analysis
generative method
autoregression
machine learning
spectral features
classification
title Combined Method for Informative Feature Selection for Speech Pathology Detection
title_full Combined Method for Informative Feature Selection for Speech Pathology Detection
title_fullStr Combined Method for Informative Feature Selection for Speech Pathology Detection
title_full_unstemmed Combined Method for Informative Feature Selection for Speech Pathology Detection
title_short Combined Method for Informative Feature Selection for Speech Pathology Detection
title_sort combined method for informative feature selection for speech pathology detection
topic voice analysis
generative method
autoregression
machine learning
spectral features
classification
url https://doklady.bsuir.by/jour/article/view/3689
work_keys_str_mv AT dslikhachov combinedmethodforinformativefeatureselectionforspeechpathologydetection
AT mivashkevich combinedmethodforinformativefeatureselectionforspeechpathologydetection
AT napetrovsky combinedmethodforinformativefeatureselectionforspeechpathologydetection
AT esazarov combinedmethodforinformativefeatureselectionforspeechpathologydetection