Speaker verification using adapted Gaussian mixture models
Speaker verification using adapted Gaussian mixture models
复制标题
DOI：
10.1006/dspr.1999.0361
复制
发表时间：
2000-01-01
影响因子：
2.9
通讯作者：
Dunn, RB
中科院分区：
文献类型：
--
作者：
Reynolds, DA;Quatieri, TF;Dunn, RB
在本文中，我们描述了麻省理工学院林肯实验室的高斯混合模型（GMM）为基础的说话人确认系统的主要元素，成功地用于几个NIST的说话人识别评估（SRE）。该系统是建立在似然比测试验证，使用简单但有效的Gestival的似然函数，一个通用的背景模型（UBM）的替代扬声器表示，和贝叶斯自适应的形式，从UBM扬声器模型。还描述和讨论了手机检测器和分数归一化的开发和使用，以大大提高验证性能。最后给出了在NIST SRE语料库上的典型性能测试和系统行为实验。(C)北京大学出版社.
In this paper we describe the major elements of MIT Lincoln Laboratory's Gaussian mixture model (GMM)-based speaker verification system used successfully in several NIST Speaker Recognition Evaluations (SREs). The system is built around the likelihood ratio test for verification, using simple but effective GMMs for likelihood functions, a universal background model (UBM) for alternative speaker representation, and a form of Bayesian adaptation to derive speaker models from the UBM. The development and use of a handset detector and score normalization to greatly improve verification performance is also described and discussed. Finally representative performance benchmarks and system behavior experiments on NIST SRE corpora are presented. (C) 2000 Academic Press.