摘要
众所周知,训练和测试环境的不同严重影响了语音识别系统的性能。该文提出了一种新的测试环境自适应的方法,它能在测试进行过程中逐步地学得环境特征,而不需要事先获得测试环境的样本数据,从而改变了语音识别系统性能。
It is well known that differences between training and testing environments seriously affect speech recognition accuracy.In this paper,a method of testing environment adaptation is proposed,which gradually learns the features of testing environment as the testing procedure,and does not need to get some testing samples beforehand,therefore it improves the performance of speech recognition system.
出处
《计算机工程与应用》
CSCD
北大核心
2002年第1期69-70,77,共3页
Computer Engineering and Applications
基金
教育部留学回国人员科研启动基金
哈尔滨工业大学跨学科交叉研究基金(编号:HIT.MD.200001)
关键词
语音识别
环境自适应
最小分类错误
speech recognition,environment adaptation,Minimum Classification Error
