SHERP

VLSI for 5000-word continuous speech recognition

Cited 0 time in webofscience Cited 0 time in scopus
Authors
Choi, Young-kyu; You, Kisun; Choi, Jungwook; Sung, Wonyong
Issue Date
2009-04
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Citation
IEEE International Acoustics, Speech and Signal Processing (ICASSP), pp. 557-560, Apr. 2009
Keywords
speech recognitionvery-large-scale integrationLVCSR
Abstract
We have developed a VLSI chip for 5,000 word speakerindependent continuous speech recognition. This chip employs a context-dependent HMM (hidden Markov model) based speech recognition algorithm, and contains emission probability and Viterbi beam search pipelined hardware units. The feature vector for speech recognition is computed using a host processor in software in order to adopt various enhancement algorithms. The amount of internal SRAM size is minimized by moving data out to the external DRAM, and a custom DRAM controller module is designed to efficiently read and write consecutive data. The experimental result shows that the implemented system has a realtime factor of 0.77 and 0.55 using SDRAM and DDR SDRAM, respectively.
ISSN
1520-6149
Language
English
URI
http://www.icassp09.com

http://hdl.handle.net/10371/6174
DOI
https://doi.org/10.1109/ICASSP.2009.4959644
Files in This Item:
Appears in Collections:
College of Engineering/Engineering Practice School (공과대학/대학원)Dept. of Electrical and Computer Engineering (전기·정보공학부)Journal Papers (저널논문_전기·정보공학부)
  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse