Subject Disentanglement Neural Network for Speech Envelope Reconstruction from EEG

TL;DR

This paper introduces SDN-Net, a neural network that disentangles subject identity from EEG signals to improve speech envelope reconstruction across different subjects, addressing variability and artifacts.

Contribution

The paper presents a novel neural network architecture with modules for disentangling subject identity, enhancing cross-subject speech envelope reconstruction accuracy.

Findings

SDN-Net outperforms state-of-the-art methods in speech envelope reconstruction.

Effective disentanglement of subject identity improves cross-subject generalization.

The approach reduces the impact of EEG variability and artifacts.

Abstract

Reconstructing speech envelopes from EEG signals is essential for exploring neural mechanisms underlying speech perception. Yet, EEG variability across subjects and physiological artifacts complicate accurate reconstruction. To address this problem, we introduce Subject Disentangling Neural Network (SDN-Net), which disentangles subject identity information from reconstructed speech envelopes to enhance cross-subject reconstruction accuracy. SDN-Net integrates three key components: MLA-Codec, MPN-MI, and CTA-MTDNN. The MLA-Codec, a fully convolutional neural network, decodes EEG signals into speech envelopes. The CTA-MTDNN module, a multi-scale time-delay neural network with channel and temporal attention, extracts subject identity features from EEG signals. Lastly, the MPN-MI module, a mutual information estimator with a multi-layer perceptron, supervises the removal of subject identity information from the reconstructed speech envelope. Experiments on the Auditory EEG Decoding Dataset demonstrate that SDN-Net achieves superior performance in inner- and cross-subject speech envelope reconstruction compared to recent state-of-the-art methods.