Fast computation of Observation Probability for Speaker-Independent Real-Time Speech Recognition 


Vol. 30,  No. 9, pp. 907-912, Sep.  2005


PDF
  Abstract

An efficient method for calculation of observation probability in CDHMM(Continous Density Hidden Markov Model) is proposed in this paper. the proposed algorithm, called FCOP(Fast Computation of Observation Probability), approximate observation probabilities in CDHMM by eliminating insignificant PDFs(Probability Density Functions) and reduces the computational load. When applied to a speech recognition system, the propsed FCOP algorithm can reduce the instruction cycles by 20%-30% and can also increase the recognition speed about 30% while minimizing the loss in its recognition rate. When implemented on a practical cellular phone, the FCOP algorithm can increase its recognition speed about 30% while suffering 0.2% loss in recognition rate.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

D. Park and J. Ahn, "Fast computation of Observation Probability for Speaker-Independent Real-Time Speech Recognition," The Journal of Korean Institute of Communications and Information Sciences, vol. 30, no. 9, pp. 907-912, 2005. DOI: .

[ACM Style]

Dong-Chul Park and Ju-Won Ahn. 2005. Fast computation of Observation Probability for Speaker-Independent Real-Time Speech Recognition. The Journal of Korean Institute of Communications and Information Sciences, 30, 9, (2005), 907-912. DOI: .

[KICS Style]

Dong-Chul Park and Ju-Won Ahn, "Fast computation of Observation Probability for Speaker-Independent Real-Time Speech Recognition," The Journal of Korean Institute of Communications and Information Sciences, vol. 30, no. 9, pp. 907-912, 9. 2005.

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880