Knowledge Distilled Ensemble Model for sEMG-based Silent Speech Interface

TL;DR

This paper introduces a lightweight ensemble deep learning model that improves sEMG-based silent speech recognition, enabling accurate classification of phonetic alphabets and potential practical, portable speech interfaces.

Contribution

The paper presents a novel knowledge-distilled ensemble deep learning model for sEMG-based silent speech interfaces, addressing previous limitations of small vocabularies and manual feature extraction.

Findings

Achieved 85.9% test accuracy on a 26 phonetic alphabet dataset.

Demonstrated the model's potential for portable silent speech systems.

Validated the effectiveness of the end-to-end approach.

Abstract

Voice disorders affect millions of people worldwide. Surface electromyography-based Silent Speech Interfaces (sEMG-based SSIs) have been explored as a potential solution for decades. However, previous works were limited by small vocabularies and manually extracted features from raw data. To address these limitations, we propose a lightweight deep learning knowledge-distilled ensemble model for sEMG-based SSI (KDE-SSI). Our model can classify a 26 NATO phonetic alphabets dataset with 3900 data samples, enabling the unambiguous generation of any English word through spelling. Extensive experiments validate the effectiveness of KDE-SSI, achieving a test accuracy of 85.9\%. Our findings also shed light on an end-to-end system for portable, practical equipment.