Abstract
In a continuous speech recognizer, the recognition problem is usually modelled as a search for the best path in a network of transitions between states. A full search can be very expensive in terms of computation and storage requirements. By adopting a segment based rather than a frame based approach,one can already obtain a severe reduction of mese requirements, but this may not be sufficient to allow for real time recognition. For our segment based Neural Network / Dynamic Programming hybrid [1], we have therefore introduced a heuristic search method. It is shown that this search is significantly faster than a Viterbi beam search, while there is no degradation of the recognition results.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
[1] J.P. Martens, A. Vorstermans, N. Cremelie (1993), “A new Dynamic Programming / Multi-Layer Perceptron Hybrid for continuous speech recognition”, in Proceedings of EUROSPEECH-93, Berlin.
[2] N J. Nilsson (1980), Principles of Artificial Intelligence, Palo Alto, California: Tioga.
[3] H. Ney (1992), “A comparative Study of Two Search Strategies for Connected Word Recognition: Dynamic Programming and Heuristic Search”, in IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 14, No. 5, May 1992.
[4] D.B. Paul (1992), “An Efficient A* Stack Decoder Algorithm for Continuous Speech Recognition with a Stochastic Language Model,” in Proceedings of ICASSP-92, San Francisco.
[5] S. Austin, R. Schwartz and P. Placeway (1991), “The Forward-Backward Search Algorithm,” in Proceedings of ICASSP-91 Toronto.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Cremelie, N., Martens, JP. (1995). Heuristic Search Methods for a Segment Based Continuous Speech Recognizer. In: Ayuso, A.J.R., Soler, J.M.L. (eds) Speech Recognition and Coding. NATO ASI Series, vol 147. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-57745-1_6
Download citation
DOI: https://doi.org/10.1007/978-3-642-57745-1_6
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-63344-7
Online ISBN: 978-3-642-57745-1
eBook Packages: Springer Book Archive