Artifact removal algorithm for an EMG-based Silent Speech Interface

Author

Wand, Michael ; Himmelsbach, Adam ; Heistermann, T. ; Janke, Matthias ; Schultz, Tanja

Author_Institution

Cognitive Syst. Lab., Karlsruhe Inst. of Technol., Karlsruhe, Germany

fYear

2013

fDate

3-7 July 2013

Firstpage

5750

Lastpage

5753

Abstract

An electromygraphic (EMG) Silent Speech Interface is a system which recognizes speech by capturing the electric potentials of the human articulatory muscles, thus enabling the user to communicate silently. This study deals with improving the EMG signal quality by removing artifacts: The EMG signals are captured by electrode arrays with multiple measuring points. On the resulting high-dimensional signal, Independent Component Analysis is performed, and artifact components are automatically detected and removed. This method reduces the Word Error Rate of the silent speech recognizer by 9.9% relative on a development corpus, and by 13.9% relative on an evaluation corpus.

Keywords

biomedical electrodes; electromyography; independent component analysis; medical signal processing; speech recognition; EMG signal quality; EMG-based silent speech interface; Word Error Rate; artifact components; artifact removal algorithm; automatical detection; development corpus; electric potentials; electrode arrays; electromygraphic silent speech interface; evaluation corpus; high-dimensional signal; human articulatory muscles; independent component analysis; multiple measuring points; silent speech recognizer; speech recognition; Electrodes; Electromyography; Feature extraction; Noise; Speech; Speech recognition; Training;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society (EMBC), 2013 35th Annual International Conference of the IEEE

Conference_Location

Osaka

ISSN

1557-170X

Type

conf

DOI

10.1109/EMBC.2013.6610857

Filename

6610857