Deep Learning Approach for Voice Pathology Detection and Classification

Deep Learning Approach for Voice Pathology Detection and Classification

Vikas Mittal, R. K. Sharma
DOI: 10.4018/IJHISI.20211001.oa28
Article PDF Download
Open access articles are freely available for download

Abstract

A non-invasive cum robust voice pathology detection and classification architecture is proposed in the current manuscript. In place of the conventional feature-based machine learning techniques, a new architecture is proposed herein which initially performs deep learning-based filtering of the input voice signal, followed by a decision-level fusion of deep learning and a non-parametric learner. The efficacy of the proposed technique is verified by performing a comparative study with very recent work on the same dataset but based on different training algorithms.The proposed architecture has five different stages.The results are recorded in terms of nine (9) different classification score indices which are – mean average Precision, sensitivity, specificity, F1 score, accuracy, error, false-positive rate, Matthews Correlation Coefficient, and the Cohen’s Kappa index. The experimental results have shown that the use of machine learning classifier can get at most 96.12% accuracy, while the proposed technique achieved the highest accuracy of 99.14% in comparison to other techniques.
Article Preview
Top

1. Introduction

Speech is the most basic form of communications known between two groups of living entities, including human beings, animals, and/or birds. As one of the fastest ways to express one’s desire or to having a task performed, speech synthesis is, nevertheless, the result of a chain of complex processes.

In the eye of a specialist, speech reflects many of the speaker’s vital traits, for example, cultural or mental health condition, physical trauma or affection, sex, various sorts of emotion, and more. Suggestively, if a person’s normal voice deviates from those of the same sex-age group, then after a thorough examination, a voice pathologist may describe this to be a case of laryngitis, or other voice dysfunction. Dysfunction in voice may be attributed to a series of gradual alteration in the neurological, physical, or medicinal activities in the voice synthesis structure of the human body. In most cases, disorders such as laryngeal cancer, vocal cord cyst, fold, nodule, polyp, and unilateral nerve paralysis are due to prolonged and inappropriate usage of the vocal organ which eventually results in hoarseness in the voice. Researchers have also cited that as many as 25% of the world’s population inevitably suffer from different types of vocal disorder due to the increasing trend of unhealthy lifestyle and self-abuse (Al-nasheri, Muhammad, Alsulaiman, & Ali, 2017; Hammami et al., 2020).

Some professionals, especially teachers, singers, and religious speakers, have a higher probability of being diagnosed with the vocal nodule pathology. The underlying reason being their nature of work as they are often required to utter a series of words for quite a long duration daily, which may also lead to an abuse of their vocal cords. As more and more swollen regions accumulate in the vocal folds, stiffness in the vocal cords will increase with time. A malfunction in the vagus nerve stimulating the larynx can also cause a disorder known as unilateral nerve paralysis, which is commonly observed as hoarseness in one’s voice. Typically, an indicative breathy phonation in the voice is prominently noticed, which in most cases has been observed with symptoms such as having difficulty in swallowing, and signs of shortness-of-breath and mild cough. Fortunately, the observed condition may be reversed via expert counsel and treatment (Steffen et al., 2011).

Complete Article List

Search this Journal:
Reset
Volume 19: 1 Issue (2024)
Volume 18: 1 Issue (2023)
Volume 17: 2 Issues (2022)
Volume 16: 4 Issues (2021)
Volume 15: 4 Issues (2020)
Volume 14: 4 Issues (2019)
Volume 13: 4 Issues (2018)
Volume 12: 4 Issues (2017)
Volume 11: 4 Issues (2016)
Volume 10: 4 Issues (2015)
Volume 9: 4 Issues (2014)
Volume 8: 4 Issues (2013)
Volume 7: 4 Issues (2012)
Volume 6: 4 Issues (2011)
Volume 5: 4 Issues (2010)
Volume 4: 4 Issues (2009)
Volume 3: 4 Issues (2008)
Volume 2: 4 Issues (2007)
Volume 1: 4 Issues (2006)
View Complete Journal Contents Listing