自动化学报2012，Vol.38Issue(4)：652-658,7.DOI:10.3724/SP.J.1004.2012.00652
基于音素解码的语种识别系统联合自适应算法研究
Research on Joint Adaptation for Phonotactic Language Recognition
摘要
Abstract
For language recognition in real application, a variety of non-language sources (I.e., channel, content, etc.) will induce mismatch between training and test utterances, which affects the recognition accuracy. This paper introduces joint adaptation to deal with the mismatch problem for the phone recognition followed by vector space model (PRVSM) system. We investigate three adaptation methods in different stage of the system: 1) acoustic model adaptation using constrained maximum likelihood linear regression (CMLLR); 2) phonotactic feature adaptation using the universal N-grams; 3) adapt-SVM for the vector space model(VSM).The joint adaptation is carried out by combining these methods and significant improvements can be obtained. Experiments on the NIST LRE 2009 evaluation corpus show that there are relative decreases of 18 % ~ 23 %, 12%~20% and 5%~9% in EER for the 30s, 10s and 3s test conditions, respectively.关键词
语种识别/音素识别器后接向量空间模型/联合自适应/受约束的最大似然线性回归/支持向量机自适应Key words
Language recognition, phone recognizer followed by vector space model (PRVSM), joint adaptation, constrained maximum likelihood linear regression (CMLLR), adapt-support vector machines (SVM)引用本文复制引用
邓妍,张卫强,刘加..基于音素解码的语种识别系统联合自适应算法研究[J].自动化学报,2012,38(4):652-658,7.基金项目
国家自然科学基金(60931160443,61005019)资助 （60931160443,61005019）
