Voice of Your Brain: Cognitive Representations of Imagined Speech, Overt Speech, and Speech Perception Based on EEG

Every people has their own voice, likewise, brain signals display distinct neural representations for each individual. Although recent studies have revealed the robustness of speechrelated paradigms for efficient brain-computer interface, the distinction on their cognitive representations with practical usability still remains to be discovered. Herein, we investigate the distinct brain patterns from electroencephalography (EEG) during imagined speech, overt speech, and speech perception in terms of subject variations with its practical use of speaker identification from single channel EEG. We performed classification of nine subjects using deep neural network that captures temporalspectral-spatial features from EEG of imagined speech, overt speech, and speech perception. Furthermore, we demonstrated the underlying neural features of individual subjects while performing imagined speech by comparing the functional connectivity and the EEG envelope features. Our results demonstrate the possibility of subject identification from single channel EEG of imagined speech and overt speech. Also, the comparison of the three speech-related paradigms will provide valuable information for the practical use of speech-related brain signals in the further studies.