Deep Learning and EEG Signal Based Biometric System: An Application of Brain Computer Interface

User identification is explored through EEG-based biometric analysis using different auditory stimuli, using the distinctiveness and security of EEG signals for reliable biometric authentication. The impact on system performance due to different auditory stimuli, namely native language songs, non-native language songs, and neutral music, was evaluated using a one-dimensional convolutional neural network (1D-CNN) model for classification. The 1D-CNN model demonstrated strong performance with an average sensitivity of 0.9666±0.10, showing its ability to correctly identify individuals using auditory stimuli. Average specificity was 0.9982±0.01, indicating excellent user differentiation and low misidentifications. The average Equal Error Rate (EER) was 0.0318±0.29, confirming the system’s reliability in recognizing users based on pattern of EEG signals.