A Digit Pipelined Dynamic Time Warp Processor

Mary Jane Irwin

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

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. Our design differs from previous systolic DTW designs in that 1) all data are represented in signed-digit, base 4 format; 2) digits are passed between processing elements in a most significant digit first, digit serial fashion; and 3) 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 very 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 also easily expandable in the number of frames per word and the warp factor used to achieve the nonlinear matching.

Original languageEnglish (US)
Pages (from-to)1412-1422
Number of pages11
JournalIEEE Transactions on Acoustics, Speech, and Signal Processing
Volume36
Issue number9
DOIs
StatePublished - Sep 1988

Fingerprint

Pattern matching
Glossaries
Processing

All Science Journal Classification (ASJC) codes

  • Signal Processing

Cite this

@article{064ef5e09fe1472d937ace97a4ddd99a,
title = "A Digit Pipelined Dynamic Time Warp Processor",
abstract = "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. Our design differs from previous systolic DTW designs in that 1) all data are represented in signed-digit, base 4 format; 2) digits are passed between processing elements in a most significant digit first, digit serial fashion; and 3) 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 very 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 also easily expandable in the number of frames per word and the warp factor used to achieve the nonlinear matching.",
author = "Irwin, {Mary Jane}",
year = "1988",
month = "9",
doi = "10.1109/29.90369",
language = "English (US)",
volume = "36",
pages = "1412--1422",
journal = "IEEE Transactions on Signal Processing",
issn = "1053-587X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "9",

}

A Digit Pipelined Dynamic Time Warp Processor. / Irwin, Mary Jane.

In: IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol. 36, No. 9, 09.1988, p. 1412-1422.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Digit Pipelined Dynamic Time Warp Processor

AU - Irwin, Mary Jane

PY - 1988/9

Y1 - 1988/9

N2 - 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. Our design differs from previous systolic DTW designs in that 1) all data are represented in signed-digit, base 4 format; 2) digits are passed between processing elements in a most significant digit first, digit serial fashion; and 3) 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 very 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 also easily expandable in the number of frames per word and the warp factor used to achieve the nonlinear matching.

AB - 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. Our design differs from previous systolic DTW designs in that 1) all data are represented in signed-digit, base 4 format; 2) digits are passed between processing elements in a most significant digit first, digit serial fashion; and 3) 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 very 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 also easily expandable in the number of frames per word and the warp factor used to achieve the nonlinear matching.

UR - http://www.scopus.com/inward/record.url?scp=0024077499&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024077499&partnerID=8YFLogxK

U2 - 10.1109/29.90369

DO - 10.1109/29.90369

M3 - Article

AN - SCOPUS:0024077499

VL - 36

SP - 1412

EP - 1422

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

SN - 1053-587X

IS - 9

ER -