Abstract
This paper presents a practical real time system for mapping dynamic glove-based hand gestures into Arabic speech. Arabic Glove-Talk (AGT) is a prototype for an intelligent system implemented to solve the problem of communication between the vocally impaired and other people. Various reasons increase the difficulty of dynamic gesture recognition. Neuro-fuzzy approaches are described to overcome this difficulty. The difficult task of gesture spotting is solved using a distance-based measure. We use the 5th Glove device to capture hand gestures. The system learns to recognise a basic vocabulary of 32 gestures. The basic vocabulary is extended to 128 gestures is tested on a test set, including 640 gestures using different types of classifiers to assign an unknown gesture to the corresponding spoken Arabic word. The minimum distance classifier, the neuro-fuzzy perceptron and the 1D-self-organising feature map based classifier result in 96.25%, 97.82% and 100% correct spoken words, respectively. After training, talkers successfully produced Arabic speech at nearly 75–90 words per minute.
Similar content being viewed by others
References
Cipolla R, Okaoto Y, Kuno Y. Robust structure from motion using motion parallax. Proceedings of the 4th International Conference on Computer Vision, IEEE Press 1993: 374–382
Davis J, Shah M. Recognizing hand gestures. Proceedings of the European Conference on Computer Vision, 1, Springer-Verlag 1994; 331–340
Starner T, Pentland A. Visual recognition of American Sign Language using Hidden Markov Models. Proceedings of the International Workshop on Face and Gesture Recognition, Zurich 1995: 184–194
Maggioni M. Gesture computer — new ways of operating a computer. Proceedings of the International Workshop on Face and Gesture Recognition, Zurich 1995: 166–171
Heap T, Hogg D. Toward 3D hand tracking using a deformable model. International Conference of Face and Gesture Recognition, Killington, VT 1996: 140–145
Moghaddam B, Pentland A. Probabilistic visual learning for object recognition. Proceedings of the 5th International Conference on Computer Vision, IEEE Press 1995: 768–793
Kunch J, Huang T. Vision based hand modeling and tracking for virtual teleconferencing and telecollaboration. Proceedings of the International Conference on Computer Vision, IEEE Press 1995: 666–671
Ahmad T, Taylor C, Lanitis A, Cootes T. Tracking and recognizing hand gestures using statistical shape models. Image & Vision Computing 1997; 15: 345–352
Freeman W, Weissman C. Television control by hand gestures. Proceedings of the first IWAFGR 1995: 179–183
Pandya A, Mcy R. Pattern Recognition with Neural Networks in C++. CRC Press, 1996
Fels S. Glove-TalkII: Mapping Hand Gestures to Speech Using Neural Networks — An Approach to Building Adaptive Interfaces. PhD thesis, Computer Science Department, University of Toronto, 1994
Fels S, Hinton G. GloveTalk: A neural network inteface between a DataGlove and a speech synthesizer. IEEE Transactions on Neural Networks 1993; 4: 2–8
Wexelblat D. A feature-based approach to continuous gesture analysis. Master's thesis, MIT, 1993
Takahashi T, Kishino F. Gesture coding based in experiments with a hand gesture interface device. SIGCHI Bulletin 1991; 23(2): 67–73
Murakami K, Taguchi H. Gesture recognition using recurrent neural networks. Computer-Human Interactin '91 Conference Proceedings 1991: 237–242
Kramer J, Leifer L. The Talking Glove: A speaking aid for nonvocal deaf and deaf-blind individuals. Proceedings of RESNA 12th Annual Conference 1989: 471–472
Kramer J, Leifer LA. Talking Glove for nonverbal deaf individuals. Technical Report CDR TR 1990 0312, Centre For Design Research, Stanford University, 1990
Kramer J. Communication system for deaf, deaf-blind and nonvocal individuals using instrumented gloves. Patent 5,047,952, Virtual Technologies, 1991
Hand C, Sexton I, Mullan M. A linguistic approach to the recognition of hand gestures. In: Designing Future Interaction Conference. Ergonomics Society/IEE, April 1994
Becker DA. Sensei: A Real-Time Recognition, feedback and Training system for T'ai Chi Gestures. MSc Thesis, Massachuetts Institute of Technology, May 1997
Hallahan WI. DECtalk software: Text-to-speech technology and implementation. Digital Technical Journal 11 April 1996
Nauck D. A fuzzy perceptron as a generic model for neuro-fuzzy approaches. Proceedings Fuzzy-Systeme'94, Munich, October 1994
Tolba, AS, Abu-Rezeq AN. A self organizing feature map for automated visual inspection of textile products. Computers in Industry 1997; 32(3): 319–333
Lee C, Yangsheng Xu. Online interactive learning of gestures for human/robor interface. IEEE International Conference on Robotics and Automation, Minneapolis 1996; 4: 2982–2987
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Tolba, A.S., Abu-Rezq, A.N. Arabic glove-talk (AGT): A communication aid for vocally impaired. Pattern Analysis & Applic 1, 218–230 (1998). https://doi.org/10.1007/BF01234769
Received:
Revised:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF01234769