Finger movement identification is a crucial method for advanced interfacing, with applications ranging from innovative HCI devices to medical prosthetics. Current approaches, such as computer vision and vibration-based detection, face limitations like restricted hand movement and environmental sensitivity. This study focuses on rectifying shortcomings in existing EMG-based hand and finger gesture recognition systems, emphasizing individual finger gestures. The proposed solution involves EMG-based classifiers with fixed electrode placement and machine learning techniques. Despite existing models like ANN and SVM, challenges persist. The paper advocates for a CNN, achieving an impressive 83.2% average classification accuracy, outperforming traditional methods. This research significantly enhances EMG-based gesture recognition, particularly for individual finger movements.