Auditory Attention Decoding from EEG using Convolutional Recurrent Neural Network

TL;DR

This paper introduces a convolutional recurrent neural network (CRNN) for auditory attention decoding from EEG data, demonstrating improved accuracy for short decoding windows and enhancing model interpretability.

Contribution

The study presents a novel CRNN-based model that outperforms existing linear and deep neural network models in auditory attention decoding, especially for short time windows.

Findings

CRNN classification model achieves ~90% accuracy for 2s and 5s windows.

CRNN regression model improves accuracy by about 5% over other regression models.

Layer visualization enhances interpretability of the deep neural network.

Abstract

The auditory attention decoding (AAD) approach was proposed to determine the identity of the attended talker in a multi-talker scenario by analyzing electroencephalography (EEG) data. Although the linear model-based method has been widely used in AAD, the linear assumption was considered oversimplified and the decoding accuracy remained lower for shorter decoding windows. Recently, nonlinear models based on deep neural networks (DNN) have been proposed to solve this problem. However, these models did not fully utilize both the spatial and temporal features of EEG, and the interpretability of DNN models was rarely investigated. In this paper, we proposed novel convolutional recurrent neural network (CRNN) based regression model and classification model, and compared them with both the linear model and the state-of-the-art DNN models. Results showed that, our proposed CRNN-based classification model outperformed others for shorter decoding windows (around 90% for 2 s and 5 s). Although worse than classification models, the decoding accuracy of the proposed CRNN-based regression model was about 5% greater than other regression models. The interpretability of DNN models was also investigated by visualizing layers' weight.