The classification of EMG signals using machine learning for the construction of a silent speech interface.

Abstract

With 7.5 million people unable to speak due to various physical and mental conditions, patients are forced to use cumbersome/inefficient devices such as eye/cheek trackers. In this study, a speech aid known as a Silent-Speech-Interface (SSI) was created. This device could be used by patients with speech disorders to communicate letters in the English-alphabet voicelessly, merely by articulating words or sentences in the mouth without producing any sounds. The SSI records EMG signals from the speech system which are then classified into speech in real-time using a trained Machine Learning model. It was found that the Support Vector Machine algorithm yielded the highest SSI accuracy of 90.1% The device created measures biomedical signals and translates them into speech accurately using a Machine Learning algorithm. This study’s findings could improve the accuracy of future SSIs by identifying the most accurate algorithms for use in an SSI.

Keywords

• silent speech interface
• emg
• machine learning
• convolutional neural network
• pattern recognition
• speech aid