vix.ing · top · new · best · stats · spec

Studying the Similarity of COVID-19 Sounds based on Correlation Analysis\n of MFCC

2020/10/17 by Mohamed Bader, Bader, Mohamed, Ismail Shahin +3
Medicine · #COVID-19 diagnosis using AI

paper · pdf · doi:10.48550/arxiv.2010.08770

Abstract

Recently there has been a formidable work which has been put up from the\npeople who are working in the frontlines such as hospitals, clinics, and labs\nalongside researchers and scientists who are also putting tremendous efforts in\nthe fight against COVID-19 pandemic. Due to the preposterous spread of the\nvirus, the integration of the artificial intelligence has taken a considerable\npart in the health sector, by implementing the fundamentals of Automatic Speech\nRecognition (ASR) and deep learning algorithms. In this paper, we illustrate\nthe importance of speech signal processing in the extraction of the\nMel-Frequency Cepstral Coefficients (MFCCs) of the COVID-19 and non-COVID-19\nsamples and find their relationship using Pearson correlation coefficients. Our\nresults show high similarity in MFCCs between different COVID-19 cough and\nbreathing sounds, while MFCC of voice is more robust between COVID-19 and\nnon-COVID-19 samples. Moreover, our results are preliminary, and there is a\npossibility to exclude the voices of COVID-19 patients from further processing\nin diagnosing the disease.\n

Related