似然得分归一化及其在与文本无关说话人确认中的应用

邓浩江; 杜利民; 万洪杰

似然得分归一化及其在与文本无关说话人确认中的应用

计量
- 文章访问数: 2470
- HTML全文浏览量: 85
- PDF下载量: 883
- 被引次数: 0
出版历程
- 收稿日期: 2004-02-23
- 修回日期: 2004-07-05
- 刊出日期: 2005-07-19

Likelihood Score Normalization and Its Application in Text-Independent Speaker Verification

摘要

摘要: 该文研究了似然得分归一化方法的原理,建立了基于自适应GMM模型的说话人确认系统,并将非特定人的背景模型与特定人的cohort模型相结合,提出了混合归一化的方法。在电话语音条件下,该文比较了不同得分归一化方法对确认系统性能的影响。实验表明,在自适应GMM模型似然比得分的基础上,T-cohort与通用背景模型混合归一化能获得最佳识别效果。当错误拒绝率为5%时,该方法可以获得0.5%的错误接受率,远远低于采用通用背景模型归一化方法的2%。
- 说话人确认; 高斯混合模型; 得分归一化; 与文本无关
Abstract: In this paper, the methodology of likelihood score normalization is studied. The text-independent speaker recognition system based on the adapted Gaussion Mixture Models(GMMs) is established, and the approach to normalize scores combining speaker-independent background model and the speaker-dependent models of cohort speaker sets are proposed. The speaker verification experiments over telephone channels show that based on the likelihood ratio of adapted GMMs system, both cohort normalization and hybrid score normalization approaches can improve the verification performance of baseline system using Universal Background Model (UBM). Specially, the hybrid approach combining UBM and cohort models selected during testing (T-cohort normalization) achieve the best performance. At a miss probability of 5%, the hybrid approach using UBM and T-cohort models reduce the false alarm rate to 0.5% compared to 2% for the baseline.

HTML全文

参考文献(1)

Doddington G R, Przybocki M A, Martin A F, Reynolds D A. The NIST speaker recognition - Overview, methodology, systems,results, perspective[J].Speech Communication.2000, 31(2- 3):225-[2]Reynolds D A. The effects of handset variability on speaker recognition performance: experiments on the switchboard corpus.In: Proc. ICASSP-1996, Atlanta, USA, May 1996:113 - 116.[3]Heck L P, Weintraub M. Handset dependent background models for robust text-independent speaker recognition. In: Proc.ICASSP-1992, Munich, Germany, 1997:1071 - 1074.[4]Reynolds D A, Quatieri T F, Dunn R B. Speaker verification using adapted Gaussian mixture models[J].Digital Signal Processing.2000, 10(1 - 3):19-[5]Dunn R B, Reynolds D A, Quatieri T F. Approaches to speaker detection and tracking in conversational speech[J].Digital Signal Processing.2000, 10(1 - 3):93-[6]Ariyaeeinia A M.[J].Sivakumaran P. Analysis and comparison of score normalization methods for text-dependent speaker verification. In Proc. EUROSPEECH97, Rhodes, Greece.1997,:-[7]Rosenberg A E, et al.. The use of cohort normalized scores for speaker verification. In Proc. ICSLP-1992, Banff, Canada, Nov.1992:599 - 602.[8]Colombi J, Ruck D, Rogers S, Oxley M, Anderson T. Cohort selection and word grammar effects for speaker recognition. In Proc. IEEE Int. Conf. Acoustics, Speech, and Signal Processing,Atlanta, GA, 1996:85 - 88.[9]Higgins A, Bahler L, Porter J. Speaker verification using randomized phrase prompting[J].Digital Signal Processing.1991,1(2):89-[10]Reynolds D A. Comparison of background normalization methods for text-independent speaker verification. In Proc.EUROSPEECH97, Rhodes, Greece, 1997:963 - 966.[11]Reynolds D, Rose R. Robust text-independent speaker identification using Gaussian mixture speaker models[J].IEEE Trans. on Speech Audio Processing.1995, 3(1):72-[12]Martin A.[J].et al.. The DET curve in assessment of detection task performance. In Proc. EUROSPEECH97, Rhodos, Greece.1997,:-

施引文献

资源附件(0)

访问统计