Face recognition has been an important research topic for many years and still holds a key position in numerous applications. It is widely used in Automated Attendance System, Human Computer Interaction, and Security systems etc. With the advancement of Artificial Intelligence, Deep Learning techniques, specifically Convolutional Neural Networks (CNNs). This paper introduces a hybrid model that combines CNN with the Softmax classifier, delivering outstanding performance in facial recognition tasks. The Softmax combined with the CNN model demonstrated an impressive accuracy of 99.95%, outperforming methods such as Viola–Jones at 74.38%, PCA at 81.81%, and Haar Cascade with CNN at 93%, while nearly matching the results of LDA, which achieved 95.45%. The proposed hybrid Softmax with CNN model has given higher accuracy and faster processing speeds compared to other traditional and hybrid machine learning techniques.