Exploring auditory acoustic features for the diagnosis of the Covid-19

TL;DR

This paper presents an automatic COVID-19 detection system using auditory features from breath, cough, and speech, achieving an AUC of 86.60%, aiming for early diagnosis through audio analysis.

Contribution

It introduces a novel system combining auditory acoustic features and bi-LSTM classifiers for COVID-19 detection from audio samples.

Findings

Achieved an AUC of 86.60% on test data.

Demonstrated high complementary behavior among features.

Validated effectiveness across breathing, cough, and speech samples.

Abstract

The current outbreak of a coronavirus, has quickly escalated to become a serious global problem that has now been declared a Public Health Emergency of International Concern by the World Health Organization. Infectious diseases know no borders, so when it comes to controlling outbreaks, timing is absolutely essential. It is so important to detect threats as early as possible, before they spread. After a first successful DiCOVA challenge, the organisers released second DiCOVA challenge with the aim of diagnosing COVID-19 through the use of breath, cough and speech audio samples. This work presents the details of the automatic system for COVID-19 detection using breath, cough and speech recordings. We developed different front-end auditory acoustic features along with a bidirectional Long Short-Term Memory (bi-LSTM) as classifier. The results are promising and have demonstrated the high complementary behaviour among the auditory acoustic features in the Breathing, Cough and Speech tracks giving an AUC of 86.60% on the test set.