A lot of interest has been generated in the field of detecting Covid-19 due to its high spreading rate. These medical techniques of detecting COVID 19, like X-ray, CT scan, and Ultrasound, necessitate one-on-one contact and take a protracted time to complete. It also costs plenty of cash. So, we propose a rapid detection method of Covid-19 through cough sound samples using Convolution Neural Network. There are three essential components to the proposed technique such steps (a) extracting audio features from cough samples or Mel images, (b) creating a feature model with extracted features, and (c) analyzing the feature model. A deep neural network (CNN) was accustomed classify cough sound samples. The cough samples from healthy and COVID infected individuals are accustomed to evaluating the model’s performance.

COVID 19, Audio features, Mel spectrogram, CNN, Machine learning