As the demand for sports coverage continues to grow, media platforms are facing the challenge of providing detailed commentary for an increasing number of games. With the rise of the Internet and social media, there is an increasing demand for real-time and detailed sports commentary across various media platforms. To meet the growing demand for comprehensive and real-time sports commentary, the research proposes the use of Convolutional Neural Networks and Retrieval Augmentation Generation for generating automatic ball-by-ball commentary for cricket matches. The final commentary provides an accurate, and semantically correct statement that is true description of what has happened on the given ball, and can predict the batter's hand with an accuracy of 70%.