Browse > Article

Effective Syllable Modeling for Korean Speech Recognition Using Continuous HMM  

김봉완 (원광대학교 음성정보기술산업지원센터)
이용주 (원광대학교 전기,전자 및 정보공학부)
Abstract
Recently attempts to we the syllable as the recognition unit to enhance performance in continuous speech recognition hate been reported. However, syllables are worse in their trainability than phones and the former have a disadvantage in that contort-dependent modeling is difficult across the syllable boundary since the number of models is much larger for syllables than for phones. In this paper, we propose a method to enhance the trainability for the syllables in Korean and phoneme-context dependent syllable modeling across the syllable boundary. An experiment in which the proposed method is applied to word recognition shows average 46.23% error reduction in comparison with the common syllable modeling. The right phone dependent syllable model showed 16.7% error reduction compared with a triphone model.
Keywords
Speech recognition; Syllable modelling; Acoustic modelling; Recognition unit;
Citations & Related Records
Times Cited By KSCI : 2  (Citation Analysis)
연도 인용수 순위
1 Design and construction of korean speech database for common use /
[ B.W.Kim;S.T.Kim;T.W.Kim;Y.I.Kim;Y.J.Lee ] / Proc. ICSP 97
2 인식 단위로서의 한국어 음절에 대한 연구 /
[ 김유진;김회린;정재호 ] / 한국음향학회지   과학기술학회마을
3 /
[ KBS ] / 표준 한국어발음대사전
4 Copernicus and the ASR challenge-waiting for kepler /
[ H.Boulard;H.Hermansky;N.Morgan ] / Proc. DARPA Speech Recognition Workshop
5 음절을 기반으로 한 한국어 음성인식 /
[ 이영호;정궁 ] / 전자공학회 논문집   과학기술학회마을
6 Syllable-based large vocaulary continuous speech recognition /
[ Aravind Ganapathiraju;Jonathan Hamaker;Joseph Picone;Mark Ordowski;George R. Doddington ] / IEEE Trans. on Speech and Audio Processing   ScienceOn
7 /
[ K.S.Choi ] / KAIST 언어자원 2001년도판, 과학기술부 핵심 소프트웨어 과제 결과물 1995-2000