Proceedings of the Korean Institute of Communication Sciences Conference (한국통신학회:학술대회논문집)

The Korean Institute of Commucations and Information Sciences (한국통신학회)
권호 표지

포르만트 주파수를 이용한 한국어 음성의 자동인식에 관한 연구

  • 김순협 (광운대학교 전자계산기공학과) ;
  • 박규태 (연세대학교 전자계산기공학과)
  • Published : 1983.04.01
Download PDF
⟨ Previous Next ⟩

Abstract

In Speech signal processing, ARMA spectral estimation method is used. It has been demonstrated that the ARMA model provides better spectral estimation then the more specialized AR model and MA model. Dynamic program is used to achieve time algnment. Speech sound similarity is defined to be proportional to the distance seperating to sound in a vector space defined by ARMA model. AS a result, the recognition rate of 97.3% for three speaker is obtained.

Keywords