Search research articles

Search research articles

Related Experiment Videos

Myo-electric signals to augment speech recognition.

A D Chan¹, K Englehart, B Hudgins

¹Institute of Biomedical Engineering, University of New Brunswick, Fredericton, Canada. biomed@unb.ca

Medical & Biological Engineering & Computing

|August 29, 2001

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Return to work after occupational injury and upper limb amputation.

Occupational medicine (Oxford, England)·2017

Same author

Classification of hand and wrist tasks of unknown force levels using muscle synergies.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2016

Same author

A comparison of classification based confidence metrics for use in the design of myoelectric control systems.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2016

Same author

On the robustness of EMG features for pattern recognition based myoelectric control: a multi-dataset comparison.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2015

Same author

A characterization of the effect of limb position on EMG features to guide the development of effective prosthetic control schemes.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2015

Same author

A multi-modal approach for hand motion classification using surface EMG and accelerometers.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2012

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Surface myo-electric signals from facial muscles show potential for improving speech recognition. This technology could enhance accuracy in noisy environments like aircraft cockpits.

Area of Science:

Biomedical Engineering
Signal Processing
Human-Computer Interaction

Background:

Conventional speech recognition struggles in noisy environments.
Myo-electric signals from facial muscles offer a potential complementary data source.
Previous research has explored myo-electric signals for control, but less for speech augmentation.

Purpose of the Study:

To investigate the presence of speech information in facial myo-electric signals.
To assess the feasibility of using these signals to augment speech recognition systems.
To evaluate performance in a simulated noisy environment.

Main Methods:

Recorded five surface myo-electric signals using Ag-AgCl electrodes on a pilot oxygen mask.
Utilized an acoustic channel for signal segmentation.

Related Experiment Videos

Processed signals with wavelet transform features and classified using linear discriminant analysis.

Tested with a ten-word vocabulary ('zero' to 'nine') in two experiments.

Main Results:

Classification errors ranged from 0.0% to 6.1% in the first experiment (non-randomized vocabulary).
Classification errors were 2.7% and 10.4% in the second experiment (randomized vocabulary).
Demonstrated significant speech information content within myo-electric signals.

Conclusions:

Surface myo-electric signals possess substantial speech information.
There is excellent potential for using these signals to enhance conventional speech recognition.
This approach could significantly improve system performance in acoustically challenging conditions.