Physiological Noise Augmentation Improves Non-Invasive Brain-to-Speech
2026-07-06 • Machine Learning
Machine Learning
AI summaryⓘ
The authors address the challenge of decoding speech signals from non-invasive brain recordings, which are noisy and less accurate than invasive methods. They introduce a technique called physiological noise augmentation (PNA), which adds realistic brain noise patterns to training data to help the decoder ignore irrelevant signals like eye or heart activity. By testing on imagined digit data, they show that PNA improves decoding accuracy when combined with averaging multiple trials. Their work suggests that making decoders aware of brain noise and using trial averaging together can enhance brain-to-speech decoding performance.
brain-to-speech decodingnon-invasive brain recordingsmagnetoencephalography (MEG)electroencephalography (EEG)physiological noisedata augmentationindependent component analysis (ICA)aniso tropic regularizationdecoding accuracybrain-computer interfaces (BCI)
Authors
Benjamin Ballyk, Teyun Kwon, Miran Özdogan, Oiwi Parker Jones
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.