Dear Members,
Can sound samples serve as acoustic biomarkers of COVID-19? If yes, an acoustics based COVID-19 diagnosis can provide a fast, contactless, inexpensive, and scalable screening scheme, with potential to supplement the existing molecular testing methods, such as RT-PCR and RAT. The DiCOVA Challenge Series is an exploration of ideas to find answers to this question.
We are excited to share with you the announcement of the "Second DiCOVA Challenge on Diagnosing COVID-19 using Acoustics". The challenge features the following three parallel tracks and one fusion track.
**********
Track-1: Breathing Sound Samples
Track-2: Cough Sound Samples
Track-3: Speech Sound Samples
Track-4: Fusion. Here participants can use any combination of above sound categories.
**********
The task in all the tracks requires building a binary classifier for COVID-19 detection using only sound samples. As was the case in the first DiCOVA challenge, a dataset and a baseline system will be released, and a leaderboard style platform for evaluation using a blind test set will be provided to all participants.
**********
Event website:
https://dicovachallenge.github.io/Registration Details:
https://dicovachallenge.github.io/#registerFlyer:
https://dicovachallenge.github.io/docs/Second_DiCOVA_challenge2022_flyer.pdfWe look forward to your participation!
The challenge timeline is chosen to suit the paper submission deadline (1st Oct. 2021) of the International Conference on Acoustics, Speech, Signal Processing (ICASSP), a flagship conference of the IEEE (
https://2022.ieeeicassp.org/).
**********
In case you would like to join hands in our dataset building effort for respiratory diagnosis, please feel free to contact us.
Cheers,
Neeraj Sharma
On behalf of Team DiCOVA