QQ客服 官方微博 反馈留言

	Channel compensation technique HNSSM for speaker recognition
	Zhou Jingfang ; Chen Yining ; Liu Jia ; Liu Runsheng
	2010-05-06 ; 2010-05-06
关键词	Practical Theoretical or Mathematical/ error statistics speaker recognition speech processing speech synthesis telecommunication channels/ channel compensation technique speaker recognition systems handset type mismatch performance degradation handset normalization synthesized speak model speaker model synthesis National Institute of Standards and Technology cepstrum mean substraction speech signal processing/ B6130E Speech recognition and synthesis B0240Z Other topics in statistics C1250C Speech recognition C1140Z Other topics in statistics C5260S Speech processing techniques
中文摘要	Handset type mismatch between the training and test speech segments causes significant performance degradation in speaker recognition systems. This paper presents channel compensation approach HNSSM (handset normalization in synthesized speak model) based on speaker model synthesis and handset normalization. Experiments using the 1999 NIST (National Institute of Standards and Technology) speaker recognition evaluation (SRE) corpus showed that the approach yielded better performance than either of the two techniques alone. Around 39. 4% relative improvement in equal error rate and 20.9% in minimum detection cost was obtained compared with the baseline system using cepstrum mean substraction (CMS).
语种	中文 ; 中文
出版者	Tsinghua Univ. Press ; China
内容类型	期刊论文
源URL	[http://hdl.handle.net/123456789/11087]
专题	清华大学
推荐引用方式 GB/T 7714	Zhou Jingfang,Chen Yining,Liu Jia,et al. Channel compensation technique HNSSM for speaker recognition[J],2010, 2010.
APA	Zhou Jingfang,Chen Yining,Liu Jia,&Liu Runsheng.(2010).Channel compensation technique HNSSM for speaker recognition..
MLA	Zhou Jingfang,et al."Channel compensation technique HNSSM for speaker recognition".(2010).

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。