Browse > Article

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

Park Dong-Chul (명지대학교 정보공학과 지능컴퓨팅 연구실)
Ahn Ju-Won (명지대학교 정보공학과 지능컴퓨팅 연구실)
Publication Information
The Journal of Korean Institute of Communications and Information Sciences / v.30, no.9C, 2005 , pp. 907-912 More about this Journal
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 obsewation probabilities in CDHMM by eliminating insignificant PDFs(Probability Density Functions) and reduces the computational load. When applied to a speech recognition system, the proposed 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.
Keywords
Speech Recognition; CDHMM; Observation Probability. PDF;
Citations & Related Records
연도 인용수 순위
  • Reference
1 K. Shinoda. and K. Iso, 'Efficient reduction of gaussian components using MDL criterion for HMM-based speech recognition,' Proc. ICASSP-02,, pp 869-872, 2002
2 L. R. Rabiner, B. H. Juang. Fundamentals of speech recognition. Prentice-Hall Inc., 1993
3 S. Ortmanns et. al. 'An efficient decoding method for real time speech recognition,' Proc. of ESCA, Eurospeech99, pp.499-502, 1999
4 T. Watanabe et. al. 'High speed speech recognition using tree structured probability density function,' Proc. ICASSP-95, vol.1, pp 556-559, 1995
5 S. Melnikoff and S. Quigley, 'Implementing log-add algorithm in hardware,' Electronics Letters, V. 39, No. 12, pp. 939-940, 2003   DOI   ScienceOn
6 S. Phadke et. al. 'On design and implementation of an embedded automatic speech recognition system,' Proc. of Int. Conf. on VLSI Design 2003, pp. 127-132, 2004   DOI
7 S. Renals, 'Phone deactivation pruning in large vocabulary continuous speech recognition,' IEEE Signal Processing Letters, vol. 3, no. 1, 1996
8 F. Elmisery et. al. 'A FPGA-based Viterbi algorithm implementation for speech recognition system,' Proc. of ICASSP-01, pp. 1217-1200, 2001