Skip to content
Open access

Efficient EEG Imagined Speech Decoding Using Channel Selection and Residual Attention-Based Lightweight Model

Nov 2026 · IEEE Sensors Journal · Vol 26, pp. 33078-33091 · 0 citations · 43 references

TL;DR

A novel decoding framework, ERP-guided dynamic wavelet phase-statistical residual network (ERP-DWPS-ResNet), is introduced that combines ERP-guided channel selection, dynamic wavelet (DW)-based signal augmentation, phase-statistical feature fusion, and a lightweight residual attention-based deep learning model for classification.

Abstract

Imagined speech is the internal articulation of words without vocalization. It offers a path for developing silent communication systems through noninvasive brain–computer interfaces (BCIs). The electroencephalographic (EEG) signals are high-dimensional, noisy, and nonstationary, making classification challenging. The primary issues arise from processing data from multiple EEG channels, many of which may contain irrelevant information. Various channel selection strategies have been investigated. Event-related potential (ERP)-based, statistical, entropy-based, spectral, and connectivity-driven approaches reduce computational overhead and enhance signal relevance. A hybrid strategy combining insights from multiple domains was also explored to select the most discriminative channels. Complementing this, a comprehensive investigation of multiple phase-based feature combinations was studied, such as those integrating phase with spectral power, entropy, Hjorth parameters, statistical descriptors, coherence, and time–frequency representations. These features capture dynamic phase interactions and neural synchrony patterns essential for imagined speech. To unify these contributions, a novel decoding framework, ERP-guided dynamic wavelet phase-statistical residual network (ERP-DWPS-ResNet), is introduced. This architecture combines ERP-guided channel selection, dynamic wavelet (DW)-based signal augmentation, phase-statistical feature fusion, and a lightweight residual attention-based deep learning model for classification. For classification, a novel lightweight deep learning model named residual hierarchically converging spatial and channel attention network (Res-HCSATNet) is introduced. The proposed framework achieves classification accuracies of 85.0%, 83.0%, 84.5%, and 94.1% for long words, short words, vowels, and short–long word pairs, respectively. Compared with transformer and BiLSTM baselines, the proposed Res-HCSATNet achieves comparable or higher accuracy with only 161 191 trainable parameters and an inference latency of 0.78–0.84 ms.

Read PDF

Similar papers

Decoding Imagined Speech: A Strictly Subject-Independent Approach Using EEG

This study presents a transparent baseline investigation of a multi-class imagined speech EEG dataset under a strictly subject-independent evaluation framework, providing a strong basis for future brain-computer interface studies targeting improved cross-subject generalization in EEG-based imagined speech decoding.

Frederik Møllskov Trier, Xiao-Peng Mao, S. Puthusserypady · 0 citations
#explainable ai Open access Oct 2026

Delineating neural contributions to electroencephalogram-based speech decoding

Objective. Speech brain–computer interfaces (BCIs) have emerged as a pivotal technology in facilitating communication for individuals with speech impairments. Utilizing electroencephalography (EEG) for noninvasive speech BCIs offers an accessible and affordable solution, potentially benefiting a broader audience. Howev...

Motoshige Sato, Eri Hatakeyama, Masakazu Inoue et al. · 0 citations
Open access

Inner speech decoding from EEG

This thesis compares three standard architectures a compact convolutional network (EEGNet), an LSTM recurrent network, and a self-attention Transformer against a hybrid model that feeds a shared convolutional front end into parallel recurrent and self-attention branches and fuses them before classification.

Pranavi Nadipalli · 0 citations
Conference Oct 2026

A statistical-token adaptive temporal network for efficient motor imagery EEG analysis

Motor imagery (MI) based brain-computer interfaces are pivotal for neurorehabilitation but are often hindered by the low signal-to-noise ratio and non-stationary nature of EEG signals. To address the trade-off between feature extraction efficiency and model complexity in existing methods, this paper proposes a lightwei...

Yuan-Long Zhu, Qing Tao · 0 citations
Open access Sep 2026

A high-precision synchronization and centroid-based framework for single-trial decoding of perceived natural speech

Decoding natural speech from EEG is constrained by low signal-to-noise ratios, commonly addressed through event-related potential (ERP) grand averaging. However, this approach acts as a low-pass filter, obscuring trial-to-trial variability and high-frequency neural dynamics. To overcome these limitations, we propose an...

Cristian Y. Olivares, Norelli Schettini · 0 citations
#artificial intelligence Preprint Nov 2026

CABrain: A Correlation-Aware Brain Foundation Model for EEG Decoding

This is the first study to integrate graph-based representations into conditional positional encoding and introduce adaptive masking within EEG foundation model pretraining and demonstrates that the proposed GCN-based positional encoding and adaptive masking strategies achieve state-of-the-art or competitive performanc...

Sugil Choi, Mooseop Kim, Chiyoon Jeong · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.