Voice familiarity detection using EEG-based Brain-Computer Interface

Brain-Computer Interface (BCI) is a direct communication pathway between brain and external devices bypassing the natural pathway of nerves and muscles. BCI enables an individual to send commands to a peripheral device using his brain activity. Electroencephalogram (EEG) is the most commonly used brain signal acquisition method as it is simple, economical and portable. Feasibility of detecting familiar vs non-familiar voice signals using EEG signals has been investigated in this paper. The results show that combination of features such as mobility and complexity of the signal gives an average accuracy of 72.2% in classifying between familiar and unfamiliar voices among 8 subjects.