COVID-19 outbreak occurred from China which spreads between people through close contact of the infected person. In this pandemic, managing such a high number of patients is difficult. Already, 5.3 million people have died. For this disease, people from all over the world became interested on telemedicine. Till now lots of people solved many problems using telemedicine that’s why we are proposing Artificial intelligence voice assistant that can help whether a particular person is COVID-19 suspected or not. Our artificial voice Assistant to help people deal with this type of circumstance. We design some questionnaires that will be asked by the machine and user will answer accordingly. Then based on their answer machine will analysis and predict, whether that particular user might have COVID or not. We took a survey and collected data from different 513 peoples. We applied several machine learning algorithms like Gini Index, Random Forest, Entropy, KNN, Decision Tree. Out of that random forest provide us the highest accuracy. Throughout the research, using random forest algorithm, we had a 92.85% prediction accuracy, which was reasonable. So, our ultimate goal is during this kind of epidemic, medical voice assistant, assists people in overcoming any problems they may be experiencing.