COVID-19 has had a devastating impact not only on various health outcomes and quality of life of individuals but also on their social well-being. Understanding the emotions of people is an essential component for effective crisis management during health crises such as COVID-19. In this study, we perform an emotion analysis of English and Spanish tweets pertaining to COVID-19. The aim of this study is to discover the emotions by using transfer learning-based text analytic approaches (including BERT and BETO). We verify the emotions by uncovering major themes from the English and Spanish tweets using topic modeling. The text analytic pipeline includes various phases. There is a data processing phase, an LDA-based topic modeling phase, followed by an emotion analysis phase. Per Figures 1 and 2, we found that the English and Spanish speakers had drastically differing emotions towards COVID-19. We show the development of emotions amongst English-speakers and Spanish-speakers in the early stages of the pandemic. We surmise that the differences between the speakers show that there are differences in conversation that lead to the emotion reactions. Moreover, we gather that the English speakers have more of a negative emotion reaction than Spanish speakers. Our analysis provides deeper insights into various issues related to psychology, public health, and economics through social media interaction during COVID-19. The knowledge discovery from our analysis can benefit health and government organizations by giving them the insight into people’s feelings, reactions, and opinions in crises. This analysis can help officials grasp the effects of public health communication and the ways communication can be improved.