Vehicle license plate recognition is a crucial application in monitoring and surveillance to enhance public safety. As the number of vehicles continues to grow, it is essential to update technologies accordingly. This approach can be applied to areas such as law enforcement, access control, and security. The state-of-the-art YOLO model has been utilized for its ability to balance accuracy and speed, which is particularly important in real-time recognition scenarios. While speed is often prioritized over accuracy in real-time recognition, the latest YOLO models successfully achieve both when trained on a custom dataset. Further the localized vehicular images are feed to Easy OCR algorithm for text detection. Score comparison techniques such as SORT algorithm is employed to finalize the text in the license plate. In this study, the model was trained on a custom dataset, achieving a mean average precision of 97.6% at a threshold of 0.5, with an average inference speed of 4 ms and recognition speeds ranging from 9 to 21 ms.