Cross-task-oriented EEG signal analysis methods: Our opinion

Attention and Sequence Modeling for Match-Mismatch Classification of Speech Stimulus and EEG Response

“…The recent success of Transformers [56] in various research areas has also made its way into the field of EEG processing. They have been used to decode the EEG signal, e.g., for motor imagery [36]- [38], [40], human brain-visual image classification [37], [39], steady-state visual evoked potential analysis [41], and emotion recognition [40].…”

“…In addition, we consider the bidirectional variants of those, referred to as BiRNN, BiLSTM, and BiGRU. Due to the success of attention-based decoding of EEG signals [36]- [41], we study to replace the spatial convolution in the EEG processing pipeline with an attention encoder (AE). In addition, the baseline model processes the speech stimulus given as a speech envelope.…”