Toward Imagined Speech based Smart Communication System: Potential Applications on Metaverse Conditions

Abstract

Recently, brain-computer interface (BCI) system is aiming to provide user-friendly and intuitive means of communication. Imagined speech has become an alternative neuroparadigm for communicative BCI since it relies directly on a person's speech production process, rather than using speechunrelated neural activity as the method of communication. Together with the current trends on BCI, we suggest a brainto-speech (BTS) system that operates by real-time decoding of multi-class imagined speech electroencephalography (EEG). Using the thirteen-class imagined speech data of nine subjects, we performed pseudo-online analysis in order to investigate the potential use of virtual BTS system in the real-world. Average accuracy of 46.54 ± 9.37 % (chance level = 7.7 %) and 75.56 ± 5.83 % (chance level = 50 %) was acquired in the thirteen-class and binary pseudo-online analysis, respectively. Together with the pseudo-online analysis of imagined speech decoding, we suggest possible form of future applications of imagined speech BCI as means of intuitive BCI communication. We provide potential applications on virtual smart home system and virtual assistant controlled by imagined speech. The virtual BTS system in this paper displays the possible form of real-world application and online training platform.