BRAIN–COMPUTER INTERFACE USING P-300 SPELLER IN TAMIL LANGUAGE

Brain–computer interface (BCI) is an interaction between the brain and the device that utilizes signals from the brain to perform certain activities that will help to communicate with the external world. In this work, the P300 Tamil speller screen has been designed using the Visual Studio software in Python. The protocol for the acquisition of electroencephalogram (EEG) signals using Recorders and Medicare Systems EEG signal processing and evaluation console — 32 Channels (RMS EEG SPEC-32) system has been devised. The EEG signals have been acquired from four different channels that include P3, P4, O1 and O2 from the participants according to the designed protocol. Following this, the data has been preprocessed and the P300 peaks have been identified. The statistical features have been extracted from the P300 peaks and classified using linear support vector machine (LSVM), Ensemble classifier and decision tree (DT). The performance metrics, F1-score, sensitivity and specificity were calculated to accurately detect the target character. This study was conducted with healthy participants; however, the success of the proposed Tamil P300 speller system shows strong potential future applicability in assistive BCI communication tools.