Browse > Article

A Variable Parameter Model based on SSMS for an On-line Speech and Character Combined Recognition System  

석수영 (영남대학교 정보통신공학과)
정호열 (영남대학교 정보통신공학과)
정현열 (영남대학교 정보통신공학과)
Abstract
A SCCRS (Speech and Character Combined Recognition System) is developed for working on mobile devices such as PDA (Personal Digital Assistants). In SCCRS, the feature extraction is separately carried out for speech and for hand-written character, but the recognition is performed in a common engine. The recognition engine employs essentially CHMM (Continuous Hidden Markov Model), which consists of variable parameter topology in order to minimize the number of model parameters and to reduce recognition time. For generating contort independent variable parameter model, we propose the SSMS(Successive State and Mixture Splitting), which gives appropriate numbers of mixture and of states through splitting in mixture domain and in time domain. The recognition results show that the proposed SSMS method can reduce the total number of GOPDD (Gaussian Output Probability Density Distribution) up to 40.0% compared to the conventional method with fixed parameter model, at the same recognition performance in speech recognition system.
Keywords
SCCRS; SSMS; Speech recognition; Character recognition; SCCRS; SSMS; Variable parameter;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Determination of the order of a markov chain by Akaike's information criterion /
[ H.Tong ] / Journal of Applied Probability   DOI   ScienceOn
2 Local Maximum방법을 이용한 가변 파라미터 CHMM 모델의 구성 /
[ 석수영;김민정;정호열;정현열 ] / 음성통신 및 신호처리 학술대회 논문집
3 한국어 음성/문자 공용인식기의 성능향상을 위한 가변 상태수 CHMM모델의 구성 /
[ 석수영;김민정;김광수;정호열;정현열 ] / 한국음향학회 학술발표대회 논문집
4 Hmm topology optimization for handwriting recognition /
[ D.Li;A.Biem;J.Subrahmonia ] / ICASSP Proc.
5 A study on HM-Nets using Decision Tree-based Successive splitting /
[ H.Takaki;K.Mashahru;I.Akinori;K.Masaki ] / ICSP-97 Proc.
6 A successive state splitting algorithm for efficient allophone modeling /
[ J.Takami;S.Sagayama ] / ICASSP-92 Proc.
7 /
[ G.Ralph;M.Stefan;W.Alex ] / Run-on recognition in an on-line handwriting recognition system
8 A successive state splitting algorithm for efficient allophone modeling /
[ J.Takami;S.Sagayama ] / ICASSP-92 Proc.
9 An on-line speech and character combined recognition system for multimodal interfaces /
[ S.Y.Suk;M.J.Kim;H.Y.Chung ] / EALPIT Proc.
10 An On-Line speech and character combined recognition system using CHMM with different model parameter /
[ S.Y.Suk;M.J.Kim;H.Y.Jung;H.Y.Chung ] / HWESPAC 8 Proc., WB32
11 A study on HM-Nets using Decision Tree-based Successive splitting /
[ H.Takaki;K.Mashahru;I.Akinori;K.Masaki ] / ICSP-97 Proc.
12 A statistical approach with HMMs for on-line cursive hangul(Korean Script) recognition /
[ B.K.Sin;J.Kim ] / Second International Conference on Document Analysis and Recognition Proc.
13 A connected spoken word recognition method by O(n) dynamic programming pattern matching algorithm /
[ S.Nakagawa ] / ICASSP Proc.