Coronavirus is a devastating virus that has killed thousands and infected millions all around the world. Being officially declared a pandemic by World Health Organization, it can quickly spread from one person to another. Therefore, to control the spread of this virus, there is a dire need for an automated system that can detect it with high accuracy without visiting a medical clinic. To obtain quick and accurate COVID testing results, which is still a critical issue research work has been conducted under this thesis where different methodologies using Artificial Intelligence Convolutional Resolution Neural Network in specific have been examined to provide efficient identification of coronavirus from cough waves. The study uses a pre-trained deep learning algorithm that aims to build a reliable solution to automatically detect and differentiate between a COVID cough and a normal cough. The dataset used in this study is the audio cough dataset that is publicly available on Kaggle. Under this work two methodologies namely ?Mobilenetv2 and CRNN+ attention? were used for the best classification of the problem stated above. Furthermore, image augmentation was employed with MobileNetv2, CRNN+ attention, along with traditional machine learning tools such as Support Vector Machine, Binary Classifier, etc. to evaluate the accuracy of the proposed model. The results obtained show that the proposed approach provides the best prediction and classification accuracy.