Svadhyaya system for the Second Diagnosing COVID-19 using Acoustics Challenge 2021

TL;DR

This paper presents a multi-modal acoustic system for COVID-19 detection using speech, breathing, and cough data, achieving high accuracy in the DiCOVA 2021 challenge.

Contribution

It introduces a combined approach leveraging three acoustic modalities, improving COVID-19 detection performance in a challenge setting.

Findings

Achieved AUCs of 86.41, 77.60, and 84.55 in breathing, cough, and speech tracks.

Fusion of modalities reached an AUC of 85.37.

Demonstrated effective multi-modal acoustic detection of COVID-19.

Abstract

This report describes the system used for detecting COVID-19 positives using three different acoustic modalities, namely speech, breathing, and cough in the second DiCOVA challenge. The proposed system is based on the combination of 4 different approaches, each focusing more on one aspect of the problem, and reaches the blind test AUCs of 86.41, 77.60, and 84.55, in the breathing, cough, and speech tracks, respectively, and the AUC of 85.37 in the fusion of these three tracks.