Chapter

Field-Programmable Logic and Applications: The Roadmap to Reconfigurable Computing

Volume 1896 of the series Lecture Notes in Computer Science pp 495-504

Date:

Reconfigurable Computing for Speech Recognition: Preliminary Findings

  • S. J. MelnikoffAffiliated withSchool of Electronic and Electrical Engineering, University of Birmingham
  • , P. B. James-RoxbyAffiliated withXilinx, Inc
  • , S. F. QuigleyAffiliated withSchool of Electronic and Electrical Engineering, University of Birmingham
  • , M. J. RussellAffiliated withSchool of Electronic and Electrical Engineering, University of Birmingham

* Final gross prices may vary according to local VAT.

Get Access

Abstract

Continuous real-time speech recognition is a highly computationally-demanding task, but one which can take good advantage of a parallel processing system. To this end, we describe proposals for, and preliminary findings of, research in implementing in programmable logic the decoder part of a speech recognition system. Recognition via Viterbi decoding of Hidden Markov Models is outlined, along with details of current implementations, which aim to exploit properties of the algorithm that could make it well-suited for devices such as FPGAs. The question of how to deal with limited resources, by reconfiguration or otherwise, is also addressed.