ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술지 A New Distance Measure for a Variable-Sized Acoustic Model Based on MDL Technique
Cited 2 time in scopus Download 1 time Share share facebook twitter linkedin kakaostory
저자
조훈영, 김상훈
발행일
201010
출처
ETRI Journal, v.32 no.5, pp.795-800
ISSN
1225-6463
출판사
한국전자통신연구원 (ETRI)
DOI
https://dx.doi.org/10.4218/etrij.10.1510.0062
협약과제
09MS4200, 휴대형 한/영 자동통역 기술개발, 김상훈
초록
Embedding a large vocabulary speech recognition system in mobile devices requires a reduced acoustic model obtained by eliminating redundant model parameters. In conventional optimization methods based on the minimum description length (MDL) criterion, a binary Gaussian tree is built at each state of a hidden Markov model by iteratively finding and merging similar mixture components. An optimal subset of the tree nodes is then selected to generate a downsized acoustic model. To obtain a better binary Gaussian tree by improving the process of finding the most similar Gaussian components, this paper proposes a new distance measure that exploits the difference in likelihood values for cases before and after two components are combined. The mixture weight of Gaussian components is also introduced in the component merging step. Experimental results show that the proposed method outperforms MDL-based optimization using either a Kullback-Leibler (KL) divergence or weighted KL divergence measure. The proposed method could also reduce the acoustic model size by 50% with less than a 1.5% increase in error rate compared to a baseline system. © 2010 ETRI.
키워드
Acoustic modeling, Minimum description length, Optimization, Parameter reduction
KSP 제안 키워드
Acoustic modeling, Baseline system, Divergence measure, Kullback-Leibler (KL) divergence, Minimum description length (MDL) criterion, Mobile devices, Model parameter, Speech recognition system, distance measure, error rate, hidden Markov Model