Large-scale training data enhances silent speech decoding with around-ear EEG.

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: 56.6% accuracy for healthy participants, 47.3% for LIS participant with large-scale training

• Primary Outcome: Silent speech decoding accuracy using 64-word classification task

• Nnt Or Sensitivity: Online top-1 accuracy: 47.2% (healthy), 26.5% (LIS); Top-5 accuracy: 76.0% (healthy), 49.1% (LIS)

• Confidence Interval: Not reported

Clinical Application

Promising for clinical implementation with wearable around-ear EEG devices, though requires substantial training data and technical expertise Could complement existing assistive communication technologies funded through NDIS. Would require TGA approval as medical device. Integration with Australian disability services and speech pathology practices needed Individuals with locked-in syndrome or severe speech impairment who retain cognitive function

Abstract

Objective. Silent speech decoding (SSD) offers a potential communication alternative for individuals with impaired vocalization. However, conventional multi-electrode electroencephalography (EEG) or facial electromyography (EMG) systems require cumbersome preparation and are unsuitable for daily use. This study evaluates the practicality of SSD using a wearable around-ear EEG device, focusing on data scaling, cross-subject transfer, vocabulary extensibility, and online decoding performance.Approach. We collected 72 h of around-ear EEG from 24 healthy participants and one individual with incomplete locked-in syndrome (LIS) during silent, vocalized, and attempted speech, and integrated these around-ear EEG recordings with prior EMG + high-density EEG datasets, yielding 282.4 total h of training data. Using a 64-word classification task as the evaluation metric, we assessed: (1) whether larger datasets improve around-ear EEG-based SSD, (2) whether healthy-participant data supplement limited LIS-participant data despite articulatory differences, (3) transferability to unseen vocabulary, and (4) online user-interface performance.Main results. Large-scale EEG/EMG data improved SSD accuracy in both healthy participants and the LIS participant. Training on the heterogeneous dataset achieved 56.6% accuracy for healthy users and 47.3% for the LIS participant. Fine-tuning this decoder for new vocabulary increased the accuracy by 22 percentage points relative to training from scratch. Regression analysis showed that, for decoding in the LIS participant, data from the LIS participant contributed approximately four times the weight of healthy-participant data, quantifying data strategies for SSD. Online experiments achieved top-1/top-5 accuracies of 47.2%/76.0% for healthy users and 26.5%/49.1% for the LIS participant.Significance. The results indicate that lightweight, commercially feasible around-ear EEG can enable practical SSD when combined with large-scale healthy-participant data, supporting online operation. Moreover, models trained on a 64-word vocabulary facilitate decoding of a new vocabulary, providing a path toward SSD systems requiring minimal LIS-participant data. This study advances non-invasive SSD systems suitable for everyday communication.

References

1. 1.Inoue, M., Hatakeyama, E., Kita, Y., & Sasai, S. (2026). Large-scale training data enhances silent speech decoding with around-ear EEG. Journal of Neural Engineering. https://doi.org/10.1088/1741-2552/ae54d0