Abstract
In this paper, which is basically a review, three approaches to the problem of recognizing word connected speech are presented. They all use the dynamic programming technique to solve the optimization problem which occurs in matching an unknown utterance against an artificially synthesized sequence of word templates. They are compared with regard to the computation time and the memory requirement.References
[1] J. BRIDLE, M. BROWN, R. CHAMBERLAIN, An Algorithm for Connected Word Recognition, Proc. ICASSP 1982, pp. 899-902.
[2] Ch. GAGNOULET, M. COUVRAT, SERAPHINE – A Connected Words Speech Recognition System, Proc. ICASSP 1982, pp. 887-890.
[3] S. GLIŃSKI et al., The Graph Search Machine (GSM): A VLSI Architecture for Connected Speech Recognition and Other Applications, Proc. of the IEE, 75, 9, pp. 1172-1184 (1987).
[2] Ch. GAGNOULET, M. COUVRAT, SERAPHINE – A Connected Words Speech Recognition System, Proc. ICASSP 1982, pp. 887-890.
[3] S. GLIŃSKI et al., The Graph Search Machine (GSM): A VLSI Architecture for Connected Speech Recognition and Other Applications, Proc. of the IEE, 75, 9, pp. 1172-1184 (1987).
