Voice Pathology Detection System Using Machine Learning Based on Internet of Things
Akmal Hakim bin Sham, N. M. Abdul Latiff, Fahad Taha AL‐Dhief, Nurul Fariesya Suhaila Md Sazihan, Sanzida Rahman, Nor Aishah Muhammad · 2023
Due to the rising incidence of voice abnormalities in the general population, voice pathology detection is an area of study that is fast expanding and receiving a lot of interest. The current method of diagnosing voice pathology is time-consuming and relies on expert practitioners' subjective evaluations. As a result, an automatic and effective method of voice pathology identification is becoming more and more necessary. By creating an Internet of Things (IoT) based speech pathology detection system based on Mel-Frequency Cepstral Coefficients (MFCC) and Support Vector Machines (SVM) algorithms, this research intends to fill this demand. In this study, the voice signals are taken from the German Saarbrucken Voice Database (SVD). The vowel /a/ was selected for this work due to its superior outcomes. This method presents a quick and easy-to-use web-based system approach that can be used by both healthcare professionals and the general public. In addition, this paper has revealed the bright future of IoT in the field of voice pathology identification.