期刊文献+

用于说话人辨识的评分规整

Score normalization for speaker identification
下载PDF
导出
摘要 在文本无关的说话人辨识中,为了提高系统在电话语音条件下的鲁棒性,提出了将说话人确认中常用的评分规整手段用于说话人辨识中,即对测试语音通过不同话者模型的评分分别进行评分规整,为测试语音选取最接近的话者模型作为系统识别输出,有效地提高了系统性能。在NIST’031spk数据库上的说话人辨识实验表明了评分规整技术对说话人辨识的有效性。 In text-independent speaker identification,robustness is very important especially for phone speech.Score normalization which is used for speaker verification is introduced into speaker identification for the first time.Through score normalization,the output scores are more reasonable for speaker identification.Experiments on text -independent GMM speaker identification in NIST'03 1spk data show noticeable improvement.
出处 《计算机工程与应用》 CSCD 北大核心 2010年第12期133-135,共3页 Computer Engineering and Applications
基金 深圳市残联培训经费项目 深圳大学科研启动基金项目(No.200811)
关键词 评分规整 说话人辨识 高斯混合模型 统一背景模型最大后验概率估计(UBM-MAP) score normalization speaker identification Gaussian Mixture Mode(lGMM) Universal Background Model-Maximum a Posterio(rUBM-MAP)
  • 相关文献

参考文献8

  • 1Reynolds D A.Speaker identification and verification using Gaus-sian mixture speaker models[J].Speech Communication,1995,17:91-108. 被引量:1
  • 2Yao Zhi-qiang,Zhou Xi,Dai Bei-qian,et al.Directly modeling of correlation matrices for GMM in speaker identification[C]//The 18th IEEE International Conference on Pattern Recognition 2006,Hong Kong,August 20-24,2006,4. 被引量:1
  • 3Reynolds D A.Speaker verification using adapted Gaussian Mixture Models[J].Diptal Signal Processing,2000,10:19-41. 被引量:1
  • 4Reynolds D A.Comparison of background normalization methods for text-independent speaker veriftcation[C]//Proc Eurospeech 1997,Rhodes,1997:963-966. 被引量:1
  • 5Roland A.Score normalization for text-independent speaker verifi-cation systems[J].Digital Signal Processing,2000,10:42-54. 被引量:1
  • 6Sturim D E,Reynolds D A.Speaker adaptive cohort selection for Tnorm in text-independent speaker verification[C]//ICASSP,2005. 被引量:1
  • 7刘明辉,陈继旭,戴蓓蒨,李辉.基于TZ Normalization规整的话者确认阈值选取[J].数据采集与处理,2005,20(3):311-317. 被引量:3
  • 8The NIST speaker recognition evaluation home[EB/OL].http://www.nist.gov/speech/tests/spk/index.htm. 被引量:1

二级参考文献6

  • 1Reynolds D A. Comparison of background normalization methods for text-independent speaker verification[A]. Proc Eurospeech 1997[C]. Rhodes, 1997.963~966. 被引量:1
  • 2The official website for the NIST speaker recognition evaluations[EB/OL]. http://www.nist.gov/speech/tests/spk. 被引量:1
  • 3Reynolds D A. Speaker verification using adapted Gaussian mixture models[J]. Digital Signal Processing,2000,10:19~41. 被引量:1
  • 4Roland A. Score normalization for text-independent speaker verification systems[J]. Digital Signal Processing, 2000,10:42~54. 被引量:1
  • 5Martin A, Przybocki M. The NIST 1999 speaker recognition evaluation-an overview[J]. Digital Signal Processing, 2000,10:1~18. 被引量:1
  • 6Rosenberg A, Delong J, Lee C, et al. The use of cohort normalized scores for speaker recognition[A]. Proc ICSLP 1992[C]. Banff, 1992.599~602. 被引量:1

共引文献2

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