A digit pipelined dynamic time warp processor [word recognition]

A custom CMOS systolic design is presented which can achieve real-time isolated word recognition for large dictionaries. The design is based on the dynamic time warping (DTW) algorithm, an exhaustive search technique which permits nonlinear pattern matching between an unknown utterance and a reference word. This design differs from previous systolic DTW designs in that all data is represented in signed-digit, base-4 format, digits are passed between processing elements in a most-significant-digit-first, digit-serial fashion, and the algorithms are pipelined at the digit level. Using most-significant digit first data flow allows digit pipelining to succeed where conventional bit-serial pipelining has failed for the arithmetic operations required in the DTW algorithm. This allows a high degree of concurrency and a high data rate to be maintained, while the pin-out requirements are kept low. As many as 25 DTW processing elements will fit on one 128-pin chip. The VLSI DTW design presented is both flexible and modular. It is independent of the number of coefficients per frame and the precision of those coefficients. The design is easily expandable in the number of frames per word and the warp factor used to achieve the nonlinear matching. >

[1]  L. Rabiner,et al.  Performance trade‐offs in dynamic time warping algorithms for isolated word recognition , 1979 .

[2]  Mary Jane Irwin,et al.  An Overview of the Penn State Design System , 1987, 24th ACM/IEEE Design Automation Conference.

[3]  Mary Jane Irwin,et al.  Mesh Arrays and Logician: A Tool for Their Efficient Generation , 1987, 24th ACM/IEEE Design Automation Conference.

[4]  Roberto Bisiani,et al.  A hardware accelerator for speech recognition algorithms , 1986, ISCA 1986.

[5]  S. Chiba,et al.  Dynamic programming algorithm optimization for spoken word recognition , 1978 .

[6]  R. Owen A VLSI dynamic time warp processor for connected and isolated word speech recognition , 1985, ICASSP '85. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[7]  J. Mann,et al.  A wafer scale DTW multiprocessor , 1986, ICASSP '86. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[8]  Ronald W. Schafer,et al.  Digital Processing of Speech Signals , 1978 .

[9]  Mary Jane Irwin,et al.  Digit-Pipelined Arnthmetic as Illustrated by the Paste-Up System: A Tutorial , 1987, Computer.

[10]  J.-L. Gauvain,et al.  A dynamic time warp VLSI processors for continuous speech recognition , 1986, ICASSP '86. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[11]  S. S. Ali,et al.  The DTWP: An LPC-based dynamic time-warping processor for isolated word recognition , 1984, AT&T Bell Lab. Tech. J..

[12]  Alex Waibel,et al.  Comparative study of nonlinear time warping techniques in isolated word speech recognition systems , 1983 .

[13]  Robert W. Brodersen,et al.  A dynamic time warp IC for a one thousand word recognition system , 1984, ICASSP.

[14]  Hiroaki Sakoe,et al.  A Dynamic Programming Approach to Continuous Speech Recognition , 1971 .

[15]  Bryan D. Ackland,et al.  Dynamic Time Warp Pattern Matching Using an Integrated Multiprocessing Array , 1983, IEEE Transactions on Computers.

[16]  S. Levinson,et al.  Considerations in dynamic time warping algorithms for discrete word recognition , 1978 .

[17]  Aaron E. Rosenberg,et al.  Performance tradeoffs in dynamic time warping algorithms for isolated word recognition , 1980 .

[18]  Bryan D. Ackland,et al.  Array configurations for dynamic time warping , 1984 .

[19]  Roberto Bisiani,et al.  A hardware accelerator for speech recognition algorithms , 1986, ISCA '86.

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

[21]  R. Bellman Dynamic programming. , 1957, Science.

[22]  Leah H. Jamieson,et al.  Simulation of a Word Recognition System on Two Parallel Architectures , 1985, IEEE Trans. Computers.