The paper discusses the creation of an acoustic-based COVID-19 diagnosis system named ‘Cough to COVID-19’ (C2C). The system was submitted to the BHI 2023 Data Competition and won first place. C2C uses pre-processing of input signals, cough-related representation extraction leveraging Wav2vec2.0, and data augmentation to enhance diagnostic accuracy. The system classifies sound signals into positive or negative for COVID-19, primarily using cough recordings. The authors emphasize the potential of C2C to improve the accuracy of COVID-19 diagnosis via cough signals.
Publication date: 3 Nov 2023
Project Page: https://github.com/Woo-jin-Chung/BHI_2023_challenge_Audio_Alchemists
Paper: https://arxiv.org/pdf/2311.00364
