Subvocal Speech Recognition Based on EMG Signal Using Independent Component Analysis and Neural Network MLP

Author

Mendes, Jose AG ; Robson, Ricardo R. ; Labidi, Sofiane ; Barros, Allan Kardec

Author_Institution

Fed. Univ. of Maranhao -Brazil, Sao Luis

Volume

1

fYear

2008

fDate

27-30 May 2008

Firstpage

221

Lastpage

224

Abstract

The performance of speech recognition systems is commonly degraded by either speech-related disabilities or by real-world factors such as the environmentpsilas noise level and reverberation. In this work, we propose a subvocal speech recognition system based on EMG signal for subvocal acquisition, Independent Component Analysis (ICA) for feature extraction and Neural Networks for classification. We have evaluated the systempsilas performance using a vowel phonemes database. The success rate was 93,99%.

Keywords

electromyography; feature extraction; independent component analysis; neural nets; signal classification; speech recognition; EMG signal; environment noise level; independent component analysis; neural network MLP; subvocal speech recognition; vowel phonemes database; Circuits; Electromyography; Feature extraction; Filtering; Independent component analysis; Microcontrollers; Neural networks; Signal processing; Speech recognition; Testing; ICA; electromyography; neural networks; subvocal recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Image and Signal Processing, 2008. CISP '08. Congress on

Conference_Location

Sanya, Hainan

Print_ISBN

978-0-7695-3119-9

Type

conf

DOI

10.1109/CISP.2008.741

Filename

4566152