基于MAP算法的无教师讲者自适应的研究  

A Study on the Chinese Continuous Speech Recognition Unsupervised Speaker Adaptation Based on Maximum a Posteriori Algorithm

作　　者：赵力[1] 邹采荣[1] 吴镇扬[1] 

机构地区：[1]东南大学无线电工程系,江苏南京210096

出　　处：《应用科学学报》2003年第4期353-356,共4页Journal of Applied Sciences

基　　金：国家自然科学基金资助项目(60272044)

摘　　要：提出了一种基于MAP算法的连续语音识别无教师讲者自适应方法,给出了简单有效的渐进自适应公式.通过识别实验结果表明,提出的方法能利用少量的自适应数据达到满意的学习效果,是一种实用和有效的讲者自适应方法.Aiming at Chinese continuous speech recognition, a novel unsupervised speaker adaptation method based on maximum a posteriori algorithm is presented in this paper, together with a simple effective formula which can give a satisfying result with only a few data. It has been proved to be an effective method to apply.

关 键 词：MAP算法 无教师讲者自适应 汉语连续语音识别 语音识别系统 再学习 

分 类 号：TN912.34[电子电信—通信与信息系统]