摘要
说话人识别可以看作语音识别的一种,是当前研究的热点之一。阐明了说话人识别的基本概念、发展状况以及原理和方法,并介绍了一种基于矢量量化的且与文本无关的说话人识别系统。
Speaker recognition may be looked as one kind of voice recognition, and it is one of researching hotspots currently. This paper illustrates the basic conception, developmental status, principle and method of speaker recognition, and introduces a VQ-based and text-independent speaker recognition system.
出处
《计算机工程与设计》
CSCD
2002年第11期12-14,62,共4页
Computer Engineering and Design
关键词
矢量量化
说话人识别
语音识别
特征提取
计算机
voice recognition
speaker recognition
vector quantization (VQ)
character pick-up
