期刊文献+

基于最大似然线性回归矩阵的说话人识别算法研究

Research on MLLR Based Speaker Recognition Algorithm
下载PDF
导出
摘要 研究了将自适应领域的最大似然线性回归(Maximum likelihood linear regression,MLLR)变换矩阵作为特征进行文本无关的说话人识别算法.本文引入了基于统一背景模型的MLLRSV-SVM说话人识别算法,并在此基础上进行高层音素聚类以进一步提高识别性能.在采用多种信道补偿技术后,在NISTSRE2006年1训练语段-1测试语段同信道和跨信道数据库上,基于MLLR特征的系统与其他最好的系统性能接近并有很强的互补性,经过简单线性融合可以极大提高识别性能. This paper uses the maximum likelihood linear regression (MLLR) as feature for text-independent speaker recognition algorithm. We introduce a universal background model (UBM) based MLLRSV-SVM algorithm first, and then extend the algorithm to multi-class for improvement. After channel compensation, in terms of the NIST 2006 SRE lconv4w-lconv4w/mic corpus, the MLLR based system is comparable with and complementary of the state of the art systems. The performance is greatly improved by simply linear fusion.
出处 《自动化学报》 EI CSCD 北大核心 2009年第5期546-550,共5页 Acta Automatica Sinica
基金 国家高技术研究发展计划(863计划)(2006AA010101 2007AA04Z223) 国家自然科学基金委员会与微软亚洲研究院联合资助项目(60776800)资助~~
关键词 说话人识别 最大似然线性回归 支持向量机 信道补偿 Speaker recognition, maximum likelihood linear regression (MLLR), support vector machine (SVM), channel compensation
  • 相关文献

参考文献13

  • 1Reynolds D A,Quatieri T F,Dunn R B.Speaker verification using adapted Gaussian mixture models.Digital Signal Processing,2000,10(1-3):19-41
  • 2Campbell W M,Sturim D E,Reynolds D A.Support vector machines using GMM supervectors for speaker verification.IEEE Signal Processing Letters,2006,13(5):308-311
  • 3Castaldo F,Colibro D,Dalmasso E,Laface P,Vair C.Compensation of nuisance factors for speaker and language recognition.IEEE Transactions on Audio,Speech,and Language Processing,2007,15(7):1969-1978
  • 4Valr C,Colibro D,Castaldo F,Daimasso F,Laface P.Channel factors compensation in model and feature domain for speaker recognition.In:Proceedings of IEEE Odyssey:The Speaker and Language Recognition Workshop.San Juan,Puerto Rico:IEEE,2006.1-6
  • 5Campbell W M,Sturim D E,Reynolds D A,Solomonoff A.SVM based speaker verification using a GMM supervector kernel and NAP variability compensation.In:Proceedings of IEEE International Conference on Acoustics,Speech,and Signal Processing.Toulouse,France:IEEE,2006.97-100
  • 6郭武,戴礼荣,王仁华.采用UBM更新量作为支持向量机特征的说话人确认[J].清华大学学报(自然科学版),2008,48(S1):704-707. 被引量:4
  • 7Stolcke A,Ferret L,Kajarekar S,Shriberg E,Venkataraman A.MLLR transforms as features in speaker recognition.In:Proceedings of the 9th European Conference on Speech Communication and Technology.Lisbon,Portugal:International Speech and Communication Association,2005.2425-2428
  • 8Karam Z N,Campbell W M.A new kernel for SVM MLLR based speaker recognition.In:Proceedings of the 8th Conference in the Annual Series of Interspeech Events and the 10th Biennial Eurospeech Conference.Antwerp,Belgium:International Speech and Communication Association,2007.290-293
  • 9Pavel M, Petr S, Jan C, Pavel C. Phonotactic language identification using high quality phoneme recognition. In: Proceedings of the 9th European Conference on Speech Communication and Technology. Lisbon, Portugal: International Speech and Communication Association, 2005. 2237-2240
  • 10边肇祺,张学工,等.模式识别.北京:清华大学出版社,1999.

二级参考文献1

共引文献21

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部