Abstract
The recognition of spoken letters of the alphabet by machine has been long accepted as a challenging and important problem, and has been viewed as a benchmark for automatic speech recognition [1]. The alphabet vocabulary contains several highly confusable subsets such as the ‘E-set’ (‘b’, ‘c’, ‘d’, ‘e’, ‘g’, ‘p’, ‘t’ ‘v’), the ‘A-set’ (‘a’, ‘j’,’ k’), and the pair’ m’ and ‘n’. Not surprisingly, most of the recognition errors occur due to the misidentification of one of the members of these subsets. It is clear that an effective algorithm for alphabet recognition must exploit all the available discriminatory information present in the speech signal.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Rabiner L. R. and Wilpon J. G.: ‘Some performance benchmarks for isolated ward speech recognition systems’. Computer Speech and Language, 2, 343–357 (1987).
Rumelhart D. E., Hinton G. E., Williams R. J.: ‘Learning internal representations by error propagation’. in Rumelhart D. E. and McClelland J. L. (eds). Parallel Distributed Processing: Explorations in the Microstructure of Cognition. 1: Foundations. MIT Press, Cambridge, MA (1986).
Woodland P. C. and Millar W.: ‘Fixed dimensional classifiers for speech recognition’. in Wheddon C. and Linggard R. (eds.) Speech and Language Processing, Chapman and Hall, London (1990).
Chan L. W. and Fallside F.: ‘An adaptive training algorithm for back propagation networks’. Computer Speech and Language, 2, 205–218 (1987).
Woodland P. C.: ‘Weight limiting, weight quantisation and generalisation in muItilayer perceptrons’. Proc IEE int Conf on Artificial Neural Networks (1989).
GilIick L. R. and Cox S. J.: ‘Some statistical issues in the comparison of speech recognition algorithms’. Proc ICASSP-89 Glasgow, Scotland (1989).
Pieraccini R. and Billi R.: ‘Experimental comparison among data compression techniques in isolated word recognition’. Proc ICASSP-83 Boston, USA (1983).
Hinton G. E.: ‘Connectionist learning procedures’. Tech Report CMU-CS-87-115, Carnegie Mellon University (1987).
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1992 British Telecommunications plc
About this chapter
Cite this chapter
Woodland, P.C. (1992). Spoken Alphabet Recognition Using Multilayer Perceptrons. In: Linggard, R., Myers, D.J., Nightingale, C. (eds) Neural Networks for Vision, Speech and Natural Language. BT Telecommunications Series, vol 1. Springer, Dordrecht. https://doi.org/10.1007/978-94-011-2360-0_10
Download citation
DOI: https://doi.org/10.1007/978-94-011-2360-0_10
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-010-5041-8
Online ISBN: 978-94-011-2360-0
eBook Packages: Springer Book Archive