• Publication
  • 1992
  • Issue No. 5 - May
  • Abstract - A Comparative Study of Two Search Strategies for Connected Word Recognition: Dynamic Programming and Heuristic Search
 This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
A Comparative Study of Two Search Strategies for Connected Word Recognition: Dynamic Programming and Heuristic Search
May 1992 (vol. 14 no. 5)
pp. 586-595

A most successful approach to recognizing continuous speech is to model the recognition problem as one of finding an optimal path through a finite state network. A comparison of two search strategies for finding the optimal path, dynamic programming and heuristic search, is presented. The comparison is based on theoretical considerations and experimental tests on a digit string task.

[1] S. E. Levinson, "Structural methods in automatic speech recognition,"Proc. IEEE, vol. 73, no. 11, pp. 1625-1650, Nov. 1985.
[2] H. Ney, "The use of a one-stage dynamic programming algorithm for connected word recognition,"IEEE Trans. Acoust. Speech Signal Processing, vol. ASSP-32, no. 2, pp. 263-271, Apr. 1984.
[3] H. Ney, "Dynamic programming as a technique for pattern recognition," inProc. 6th Int. Conf. Pattern Recognition, (Munich, Germany), Oct. 1982, pp. 1119-1125.
[4] B. Lowerre and R. Reddy, "The HARPY speech understanding system," inTrends in Speech Recognition(W. A. Lea, Ed.). Englewood Cliffs, NJ: Prentice-Hall, 1980, pp. 340-360.
[5] N. Nilsson,Principles of Artificial Intelligence. Palo Alto, CA: Tioga, 1980.
[6] P. M. Winston,Artificial Intelligence. Reading, MA: Addison-Wesley, 1984.
[7] F. Jelinek, L. R. Bahl, and R. L. Mercer, "Design of a linguistic statistical decoder for the recognition of continuous speech,"IEEE Trans. Informat. Theory, vol. IT-21, no. 3, pp. 250-256, May 1975.
[8] M. K. Brown and L. R. Rabiner, "An adaptive, ordered, graph search technique for dynamic time warping for isolated word recognition,"IEEE Trans. Acoust. Speech Signal Processing, vol. ASSP-30, no. 4, pp. 535-543, Aug. 1982.
[9] J. C. Spohrer, P. F. Brown, P. H. Hochschild, and J. K. Baker, "Partial traceback in continuous speech recognition," inProc. Int. Conf. Cybern. Soc., (Boston), Oct. 1980, 36-42.
[10] J. S. Bridle, M. D. Brown, and R. M. Chamberlain, "A one-pass algorithm for connected word recognition," inProc. 1982 IEEE Int. Conf. Acoust. Speech Signal Processing, (Paris, France), May 1982, pp. 899-902.
[11] R. Sedgewick,Algorithms. Reading, MA, Addison-Wesley, 1983.
[12] Y. Kamp, H. Bourlard, H. Ney, and C. Wellekens: "Speaker-dependent connected speech recognition via dynamic programming and statistical methods," inSpeech and Speaker Recognition. Karger, Basel: Bibliotheca Phonetica, 1985, pp. 115-148.
[13] H. Ney, D. Mergel, A. Noll, and A. Paeseler: "A data driven organization of the dynamic programming beam search for continuous speech recognition," inProc. IEEE Int. Conf. Acoust. Speech Signal Processing(Dallas, TX), Apr. 1987, pp. 833-836.
[14] V. Steinbisset al., "A 10000-word continuous speech recognition system," inProc. 1990 Int. Conf. Acoust. Speech Signal Processing(Albuquerque, NM), Apr. 1990, pp. 57-60.

Index Terms:
connected word recognition; dynamic programming; heuristic search; continuous speech; optimal path; finite state network; dynamic programming; graph theory; heuristic programming; search problems; speech recognition
Citation:
H. Ney, "A Comparative Study of Two Search Strategies for Connected Word Recognition: Dynamic Programming and Heuristic Search," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 14, no. 5, pp. 586-595, May 1992, doi:10.1109/34.134063
Usage of this product signifies your acceptance of the Terms of Use.