Myo-electric signals to augment speech recognition

  • A. D. C. Chan
  • K. Englehart
  • B. Hudgins
  • D. F. Lovely
Communication

Abstract

It is proposed that myo-electric signals can be used to augment conventional speech-recognition systems to improve their performance under acoustically noisy conditions (e.g. in an aircraft cockpit). A preliminary study is performed to ascertain the presence of speech information within myo-electric signals from facial muscles. Five surface myo-electric signals are recorded during speech, using Ag−AgCl button electrodes embedded in a pilot oxygen mask. An acoustic channel is also recorded to enable segmentation of the recorded myo-electric signal. These segments are processed off-line, using a wavelet transform feature set, and classified with linear discriminant analysis. Two experiments are performed, using a ten-word vocabulary consisting of the numbers ‘zero’ to ‘nine’. Five subjects are tested in the first experiment, where the vocabulary is not randomised. Subjects repeat each word continuously for 1 min; classification errors range from 0.0% to 6.1%. Two of the subjects perform the second experiment, saying words from the vocabulary randomly; classification errors are 2.7% and 10.4%. The results demonstrate that there is excellent potential for using surface myo-electric signals to enhance the performance of a conventional speech-recognition system.

Keywords

Speech recognition Myo-electric signal Wavelet transform Pattern recognition Biological signal processing 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Day, S. H. (1990): ‘Recognition of speech utilizing the myoelectric signals of neck muscles — An advanced study in the time domain’. MSc thesis, Auburn UniversityGoogle Scholar
  2. Englehart, K., Hudgins, B., Parker, P.A., andStevenson, M. (1999): ‘Classification of the myoelectric signal using time-frequency based representations,’Med. Eng. Phys.,21, pp. 431–438CrossRefGoogle Scholar
  3. Hudgins, B. (1991) ‘A new approach to multifunction myoelectric control’. PhD thesis, University of New Brunswick, Fredericton, NB, CanadaGoogle Scholar
  4. Hudgins, B., Parker, P. A., andScott, R. N. (1993): ‘A new strategy for multifunction myoelectric control’,IEEE Trans. Biomed. Eng.,40, pp. 82–94CrossRefGoogle Scholar
  5. Long, D. W. (1990): ‘Speech recognition from myoelectric signals — Frequency spectrum shaping effects on speech recognition’. MSc thesis, Auburn UniversityGoogle Scholar
  6. Lovely, D. F. (1993): ‘Low noise electrode amplifier for use in evoked potential studies’. 19th Canadian Medical & Biological Engineering Society (CMBES) Conference, Ottawa, Ont., Canada, pp. 236–237Google Scholar
  7. Morse, M. S., andO'Brien, E. M. (1986): ‘Research summary of a scheme to ascertain the availability of speech information in the myoelectric signals of neck and head muscles using surface electrodes’,Comput. Biol. Med.,16, pp. 399–410CrossRefGoogle Scholar
  8. Research and Technology Organization (North Atlantic Treaty Organization) (1998): ‘Alternative control technologies’. RTO Technical Report 7Google Scholar
  9. Sugie, N., andTsunoda, K. (1985): ‘A speech prosthesis employing a speech synthesizer — vowel discrimination from perioral muscle activities and vowel production’,IEEE Trans. Biomed. Eng.,32, pp. 485–490Google Scholar
  10. White, R. G., andBeckett, P. (1983): ‘Increased aircraft survivability using direct voice input’. Proc. AGARD Flight Mechanics Panel Symposium: Flight mechanics and system design lessons from operational experience, AGARD-CP-347, AthensGoogle Scholar

Copyright information

© IFMBE 2001

Authors and Affiliations

  • A. D. C. Chan
    • 1
  • K. Englehart
    • 1
  • B. Hudgins
    • 1
  • D. F. Lovely
    • 1
  1. 1.Institute of Biomedical EngineeringUniversity of New BrunswickFrederictonCanada

Personalised recommendations