Physiological Noise Augmentation Improves Non-Invasive Brain-to-Speech

2026-07-06

AI Summary

This study addresses the high word error rates in non-invasive brain-to-speech decoding, which are primarily caused by the low signal-to-noise ratio of electroencephalography (EEG) and contamination from physiological artifacts such as eye movements and cardiac activity. To mitigate these challenges, the work introduces Physiological Noise Augmentation (PNA), a novel approach inspired by noise augmentation techniques in speech recognition. PNA leverages Independent Component Analysis (ICA) to separate clean neural components from physiological noise sources and then synthesizes label-preserving augmented samples by recombining these components in controlled proportions, thereby imposing anisotropic regularization along artifact-prone directions. Integrated with an EEGNet decoder and trial averaging, PNA achieves an absolute improvement of 4.7 percentage points in decoding accuracy on the MegNIST dataset under 10-trial averaging, substantially enhancing model robustness and overall system performance.

Abstract

Non-invasive brain-to-speech decoding aims to restore communication to patients suffering from neurodegenerative disease, without the risks of neurosurgery. Existing MEG- and EEG-based methods, while scalable, continue to suffer from high word error rates driven by relatively low signal-to-noise ratios compared to invasive recordings. We propose physiological noise augmentation (PNA), a data augmentation method that explicitly trains decoders to become invariant to task-agnostic artifacts (e.g. ocular and cardiac activity). PNA draws inspiration from automatic speech recognition systems, where environmental noise (e.g. dogs barking, city traffic) is added to clean speech to improve robustness. Analogously, we decompose brain recordings into clean data and noise artifacts using independent component analysis (ICA), before scaling and remixing to generate biophysically realistic, label-preserving training examples. We show that PNA approximates anisotropic regularization, penalizing decoder sensitivity along artifact-dominated directions. On MegNIST, a 12k-trial imagined-digit MEG dataset, PNA with 10-trial averaging improves EEGNet decoding accuracy by 4.7 percentage points (absolute) over training on real data alone. Our results suggest that artifact-aware augmentation and trial averaging are complementary tools for improving robustness in non-invasive speech BCIs.

Problem

• brain-to-speech decoding
• physiological noise
• non-invasive BCI
• signal-to-noise ratio
• word error rate

Innovation

• Physiological Noise Augmentation
• Brain-to-Speech Decoding
• Non-invasive BCI
• Data Augmentation
• Artifact Invariance