Learning Subject-Invariant Representations from Speech-Evoked EEG Using Variational Autoencoders

TL;DR

This paper introduces a novel variational autoencoder approach to extract subject-invariant features from speech-evoked EEG data, enhancing generalization and classification accuracy across subjects.

Contribution

It adapts factorized hierarchical variational autoencoders to disentangle subject and content features in EEG, improving cross-subject speech processing.

Findings

Subject accuracy reaches 98.96% on subject latent space.

Content classification accuracy reaches 62.91%.

Disentangled representations improve EEG-based speech understanding.

Abstract

The electroencephalogram (EEG) is a powerful method to understand how the brain processes speech. Linear models have recently been replaced for this purpose with deep neural networks and yield promising results. In related EEG classification fields, it is shown that explicitly modeling subject-invariant features improves generalization of models across subjects and benefits classification accuracy. In this work, we adapt factorized hierarchical variational autoencoders to exploit parallel EEG recordings of the same stimuli. We model EEG into two disentangled latent spaces. Subject accuracy reaches 98.96% and 1.60% on respectively the subject and content latent space, whereas binary content classification experiments reach an accuracy of 51.51% and 62.91% on respectively the subject and content latent space.