Multi-stream HMM for EMG-based speech recognition

H. Manabe, Z. Zhang

Research output: Contribution to journalConference article

35 Citations (Scopus)

Abstract

A technique for improving the recognition accuracy of EMG-based speech recognition by applying existing speech recognition technologies is proposed. The authors have proposed an EMG-based speech recognition system that requires only mouth movements, voice need not be generated. A multi-stream HMM (Hidden Markov Model) and feature extraction technique are applied to EMG-based speech recognition. 3 channel facial EMG signals are collected from ten subjects when uttering 10 Japanese isolated digits. One channel corresponds to one stream. By examining various features, we found that the delta component of the static parameter leads to higher accuracy. Compared to equal stream weighting, the individual optimization of stream weights increased recognition accuracy by 4. 0% which corresponds to a 12.8% reduction in error rate. This result shows that multi-stream HMM is effective for the classification of EMG.

Original languageEnglish
Pages (from-to)4389-4392
Number of pages4
JournalAnnual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings
Volume26 VI
Publication statusPublished - 2004 Dec 1
EventConference Proceedings - 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2004 - San Francisco, CA, United States
Duration: 2004 Sep 12004 Sep 5

    Fingerprint

Keywords

  • EMG
  • Multi-stream HMM
  • Speech recognition

ASJC Scopus subject areas

  • Signal Processing
  • Biomedical Engineering
  • Computer Vision and Pattern Recognition
  • Health Informatics

Cite this