Understanding effect of speech perception in EEG based speech recognition systems
This addresses the problem of making EEG-based speech recognition more robust for applications like brain-computer interfaces, though it appears incremental as it builds on prior work.
The paper investigates whether speech perception components can be separated from EEG signals to improve EEG-based speech recognition systems, achieving very low normalized RMSE in predicting EEG signals between speaking and listening conditions and improving previous CTC model results.
The electroencephalography (EEG) signals recorded in parallel with speech are used to perform isolated and continuous speech recognition. During speaking process, one also hears his or her own speech and this speech perception is also reflected in the recorded EEG signals. In this paper we investigate whether it is possible to separate out this speech perception component from EEG signals in order to design more robust EEG based speech recognition systems. We further demonstrate predicting EEG signals recorded in parallel with speaking from EEG signals recorded in parallel with passive listening and vice versa with very low normalized root mean squared error (RMSE). We finally demonstrate both isolated and continuous speech recognition using EEG signals recorded in parallel with listening, speaking and improve the previous connectionist temporal classification (CTC) model results demonstrated by authors in [1] using their data set.