Proceedings of the IEEK Conference (대한전자공학회:학술대회논문집)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지

Implementation of HMM-Based Speech Recognizer Using TMS320C6711 DSP

  • Bae Hyojoon (Electrical Engineering and Computer Science, Korea Advanced Institute of Science and Technology) ;
  • Jung Sungyun (School of Electronic and Electrical Engineering, Kyungpook National University) ;
  • Son Jongmok (School of Electronic and Electrical Engineering, Kyungpook National University) ;
  • Kwon Hongseok (School of Electronic and Electrical Engineering, Kyungpook National University) ;
  • Kim Siho (School of Electronic and Electrical Engineering, Kyungpook National University) ;
  • Bae Keunsung (School of Electronic and Electrical Engineering, Kyungpook National University)
  • Published : 2004.08.01
Download PDF
⟨ Previous Next ⟩

Abstract

This paper focuses on the DSP implementation of an HMM-based speech recognizer that can handle several hundred words of vocabulary size as well as speaker independency. First, we develop an HMM-based speech recognition system on the PC that operates on the frame basis with parallel processing of feature extraction and Viterbi decoding to make the processing delay as small as possible. Many techniques such as linear discriminant analysis, state-based Gaussian selection, and phonetic tied mixture model are employed for reduction of computational burden and memory size. The system is then properly optimized and compiled on the TMS320C6711 DSP for real-time operation. The implemented system uses 486kbytes of memory for data and acoustic models, and 24.5kbytes for program code. Maximum required time of 29.2ms for processing a frame of 32ms of speech validates real-time operation of the implemented system.

Keywords