한국어 음성 분할을 위한 특징 검출에 관한 연구

A Study on the Feature Extraction for the Segmentation of Korean Speech

  • 이극 (서울대학교 공과대학 전자계산기공학과) ;
  • 황희융 (서울대학교 공과대학 전자계산기공학과)
  • 발행 : 1987.11.20

초록

The speech recognition system usually consists of two modules, segmentation module and identification module. So, the performance of the system heavily depends on the segmentation accuracy and the segmentation unit. This paper is concerned with the agreeable features for segmentation in syllables. Total energy and two band width energy. (LE:4000-5000Hz and HE:900-3100Hz) are suitable cues for segmentation. And we testify it through the experiment using connected digit.

키워드