A Parametric Approach for Classification of Distortions in Pathological Voices


In biomedical acoustics, distortion in voice signals, commonly present during acquisition and transmission, adversely affects acoustic features extracted from pathological voice. Information on the type of distortion can help in compensating for its effects. This paper proposes a new approach to detecting four major types of commonly encountered distortion in remote analysis of pathological voice, namely background noise, reverberation, clipping and coding. In this approach, by applying factor analysis to Gaussian mixture model mean supervectors, distortions in variable-duration recordings are modeled by fixed-length, low-dimensional channel vectors. Then, linear discriminant analysis (LDA) is used to remove the remaining nuisance effects in the channel vectors. Finally, two different classifiers, namely support vector machines and probabilistic LDA classify the different types of distortion. Experimental results obtained using Parkinson's voices, as an example of pathological voice, show 11.4% relative improvement in performance over systems which directly use acoustic features for distortion classification.

Publication DOI: https://doi.org/10.1109/ICASSP.2018.8461316
Divisions: College of Engineering & Physical Sciences > Systems analytics research institute (SARI)
Additional Information: © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Event Title: 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Event Type: Other
Event Dates: 2018-04-15 - 2018-04-20
ISBN: 978-1-5386-4659-5, 978-1-5386-4658-8
Last Modified: 10 Jun 2024 07:50
Date Deposited: 21 Sep 2018 07:40
Full Text Link:
Related URLs: https://ieeexpl ... rce=SEARCHALERT (Publisher URL)
PURE Output Type: Conference contribution
Published Date: 2018-09-13
Accepted Date: 2018-01-29
Authors: Poorjam, Amir Hossein
Little, Max A (ORCID Profile 0000-0002-1507-3822)
Jensen, Jesper Rindom
Christensen, Mads Græsbøll



Version: Accepted Version

| Preview

Export / Share Citation


Additional statistics for this record