• DocumentCode
    2895501
  • Title

    VLSI architecture for HMM-based speech recognition systems and its verification platform

  • Author

    Yoshizawa, Shingo ; Wada, Naoya ; Hayasaka, Noboru ; Miyanaga, Yoshikazu

  • Author_Institution
    Graduate Sch. of Eng., Hokkaido Univ., Sapporo, Japan
  • Volume
    2
  • fYear
    2004
  • fDate
    26-29 Oct. 2004
  • Firstpage
    700
  • Abstract
    This work presents VLSI architecture in HMM-based speech recognition for high-speed operation and its verification platform to test various tasks on recognition systems. The proposed architecture effectively utilizes independent computations on the HMM structure. It can reduce processing time and/or extend the word vocabulary considerably. We designed a complete recognizer, including speech analysis and noise robustness parts, and developed a FPGA platform and a computer-aided design tool that generates source codes and testing data. The recognizer provides 35.7 μs/word in response time for word recognition tasks on a 0.18 μm CMOS technology and the FPGA platform enables real-time recognition experiments under various conditions.
  • Keywords
    CMOS logic circuits; VLSI; circuit CAD; field programmable gate arrays; hidden Markov models; integrated circuit design; integrated circuit noise; logic CAD; speech recognition equipment; 0.18 micron; 35.7 mus; CMOS technology; FPGA platform; HMM structure; HMM-based speech recognition systems; VLSI architecture; computer-aided design tool; independent computations; noise robustness; processing time; real-time recognition; response time; source codes; speech analysis; speech recognizer design; testing data; verification platform; word recognition tasks; word vocabulary; CMOS technology; Computer architecture; Field programmable gate arrays; Hidden Markov models; Noise generators; Speech analysis; Speech recognition; System testing; Very large scale integration; Vocabulary;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Communications and Information Technology, 2004. ISCIT 2004. IEEE International Symposium on
  • Print_ISBN
    0-7803-8593-4
  • Type

    conf

  • DOI
    10.1109/ISCIT.2004.1413805
  • Filename
    1413805