A Systolic FPGA Architecture of Two-Level Dynamic Programming for Connected Speech Recognition

In this paper, we present an efficient architecture for connected word recognition that can be implemented with field programmable gate array (FPGA). The architecture consists of newly derived two-level dynamic programming (TLDP) that use only bit addition and shift operations. The advantages of this architecture are the spatial efficiency to accommodate more words with limited space and the absence of multiplications to increase computational speed by reducing propagation delays. The architecture is highly regular, consisting of identical and simple processing elements with only nearest-neighbor communication, and external communication occurs with the end processing elements. In order to verify the proposed architecture, we have also designed and implemented it, prototyping with Xilinx FPGAs running at 33 MHz.

[1]  Jun-Dong Cho,et al.  Low power Viterbi search architecture using inverse hidden Markov model , 2000, 2000 IEEE Workshop on SiGNAL PROCESSING SYSTEMS. SiPS 2000. Design and Implementation (Cat. No.00TH8528).

[2]  Seiichi Nakagawa.  A connected spoken word recognition method by O(n) dynamic programming pattern matching algorithm , 1983, ICASSP.

[3]  Jun-Dong Cho,et al.  Low power VLSI architecture of Viterbi scorer for HMM-based isolated word recognition , 2002, Proceedings International Symposium on Quality Electronic Design.

[4]  Yoshikazu Miyanaga,et al.  A low-power VLSI design of an HMM based speech recognition system , 2002, The 2002 45th Midwest Symposium on Circuits and Systems, 2002. MWSCAS-2002..

[5]  Jer Min Jou,et al.  An efficient VLSI architecture for HMM-based speech recognition , 2001, ICECS 2001. 8th IEEE International Conference on Electronics, Circuits and Systems (Cat. No.01EX483).

[6]  H. Sakoe,et al.  Two-level DP-matching--A dynamic programming-based pattern matching algorithm for connected word recognition , 1979 .

[7]  Biing-Hwang Juang,et al.  Fundamentals of speech recognition , 1993, Prentice Hall signal processing series.

[8]  G.C. Cardarilli,et al.  Hardware oriented architectures for continuous-speech speaker-independent ASR systems , 2004, Proceedings of the Fourth IEEE International Symposium on Signal Processing and Information Technology, 2004..

[9]  A. H. Khalil,et al.  A FPGA-based HMM for a discrete Arabic speech recognition system , 2003, Proceedings of the 12th IEEE International Conference on Fuzzy Systems (Cat. No.03CH37442).

[10]  Fabian Vargas,et al.  A FPGA-based Viterbi algorithm implementation for speech recognition systems , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).

[11]  Hermann Ney.  A Comparative Study of Two Search Strategies for Connected Word Recognition: Dynamic Programming and Heuristic Search , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  M.G. Bellanger,et al.  Digital processing of speech signals , 1980, Proceedings of the IEEE.

[13]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[14]  Oliver Chiu-sing Choy,et al.  An HMM-based speech recognition IC , 2003, Proceedings of the 2003 International Symposium on Circuits and Systems, 2003. ISCAS '03..

[15]  Shambhu J. Upadhyaya,et al.  A parallel VLSI implementation of Viterbi algorithm for accelerated word recognition , 1993, [1993] Proceedings Third Great Lakes Symposium on VLSI-Design Automation of High Performance VLSI Systems.