Towards EMG-to-Speech with a Necklace Form Factor

TL;DR

This paper investigates a novel neck-worn EMG device for speech decoding, achieving high accuracy and revealing important electrode configurations, with potential for more convenient speech interfaces.

Contribution

It introduces a neck-based EMG device for speech decoding, demonstrating high accuracy and analyzing electrode importance and speech-EMG relationships.

Findings

92.7% classification accuracy with the neck device

More than two electrodes improve performance

Linear relationship between EMG spectrograms and speech representations

Abstract

Electrodes for decoding speech from electromyography (EMG) are typically placed on the face, requiring adhesives that are inconvenient and skin-irritating if used regularly. We explore a different device form factor, where dry electrodes are placed around the neck instead. 11-word, multi-speaker voiced EMG classifiers trained on data recorded with this device achieve 92.7% accuracy. Ablation studies reveal the importance of having more than two electrodes on the neck, and phonological analyses reveal similar classification confusions between neck-only and neck-and-face form factors. Finally, speech-EMG correlation experiments demonstrate a linear relationship between many EMG spectrogram frequency bins and self-supervised speech representation dimensions.