Speech Imagery Classification using Length-Wise Training based on Deep Learning
This research is significant for individuals seeking direct communication through brain-computer interfaces, particularly those who might benefit from speech imagery, by improving the classification performance of imagined speech from EEG signals.
This paper addresses the challenge of classifying speech imagery from EEG signals, which are known for their intricate and non-stationary properties. The authors propose a length-wise training strategy combined with a hierarchical convolutional neural network and a novel loss function, achieving competitive performance in speech imagery classification.
Brain-computer interface uses brain signals to control external devices without actual control behavior. Recently, speech imagery has been studied for direct communication using language. Speech imagery uses brain signals generated when the user imagines speech. Unlike motor imagery, speech imagery still has unknown characteristics. Additionally, electroencephalography has intricate and non-stationary properties resulting in insufficient decoding performance. In addition, speech imagery is difficult to utilize spatial features. In this study, we designed length-wise training that allows the model to classify a series of a small number of words. In addition, we proposed hierarchical convolutional neural network structure and loss function to maximize the training strategy. The proposed method showed competitive performance in speech imagery classification. Hence, we demonstrated that the length of the word is a clue at improving classification performance.