Abstract
This paper is based on classification of various voice disorders using non-invasive methods with the help of Machine Learning algorithms. In this work, voice samples for three disorders — Dysphonia, Vocal Fold Paralysis and Laryngitis along with normal speech samples are considered. A comprehensive database for each category (4 classes) is created. Using speech processing and feature extraction techniques, the relevant features are extracted, which are stored in a supervector using GMM-UBM. This supervector is then projected on a low dimensional feature vector known as ‘i-Vectors’ using total variability factor analysis. These extracted features are stored and a feature matrix is created. The parameters obtained hereafter are used to train the system using Support Vector Machines, Naive Bayes and K-NN. The trained system is capable of classification of voice disorders. The accuracy using the above-mentioned classifiers is in the range of 84% to 96%.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.