期刊文献+

用于语音识别的空间相关性变换 被引量:2

Spatial correlation transformation for speech recognition
原文传递
导出
摘要 针对经典隐含Markov模型忽略了语音信号之间的依存关系的问题,提出一种线性特征变换——空间相关性变换,利用同一个说话人的不同语音单元之间的相关性(空间相关性)得到鉴别性能更好的新特征。该变换的最优变换矩阵在最小协方差准则下得到。识别系统采用新特征及其模型参数代替原特征及其模型参数进行Viterbi搜索。实现空间相关性变换的关键是最优变换矩阵的计算,提出了两种相应的算法。实验结果表明:该方法在说话人无关识别系统上取得了比自适应方法更好的性能,同时该方法与自适应方法结合应用可进一步提高系统性能。 The traditional Hidden Markov model for speech recognition ignores the relationships between speech signals. This paper presents a linear feature transformation, Spatial Correlation Transformation, to utilize the correlation between different acoustic units of the same speaker (Spatial Correlation) to obtain new features having better discrimination. The optimum transformation matrix is determined based on the Minimum Covariance criterion. The recognition system uses these new features and the corresponding model parameters in the Viterhi search instead of the original features. The key to the transformation is the calculation of the optimum transformation matrix. Experiments show that this approach achieves better performance than adaptation approaches on the speaker independent recognition system. Moreover, the combination of this approach and adaptation approaches further improves the system performance.
出处 《清华大学学报(自然科学版)》 EI CAS CSCD 北大核心 2009年第10期1655-1659,共5页 Journal of Tsinghua University(Science and Technology)
关键词 语音识别 空间相关性 特征变换 最小协方差 speech recognition spatial correlation feature transformation minimum covariance
  • 相关文献

参考文献6

  • 1Young S. Statistical modelling in continuous speech recognition [C]// Proc International Conference on Uncertainty in Artificial Intelligence, Seattle, USA, 2001.
  • 2Leggetter C J, Woodland P C. Maximum likelihood linear regression for speaker adaptation of continuous density hidden markov models [J]. Computer Speech and Language, 1995, 9(2): 171-185.
  • 3Gales M J F, Woodland P C. Mean and variance adaptation within the MLLR framework [J]. Computer Speech and Language, 1996, 10(4) : 249 - 264.
  • 4Hazen T. The use of speaker correlation information for automatic speech recognition [D]. Cambridge: Mass, Inst, Technol, Jan 1998.
  • 5Kuhn R, Junqua J C, Nguyen P, et al. Rapid speaker adaptation in eigenvoice space [J]. IEEE Trans on Speech and Audio Processing, 2000, 8(6) : 695 - 707.
  • 6王作英.基于段长分布的HMM语音识别模型[C].见:第二届全国汉字语音识别会议,庐山,1989

共引文献1

同被引文献15

  • 1Leggetter C J and Woodland P C. Maximum likelihood linear regression for speaker adaptation of continuous density hidden markov models. Computer Speech and Language, 1995, 9(2): 171-185.
  • 2Kuhn R, Junqua J C, and Nguyen P, et al.. Rapid speaker adaptation in eigenvoice space. IEEE Transactions on Speech and Audio Processing, 2000, 8(6): 695-707.
  • 3Anastasakos Tasos, McDonough John, and Makhoul John. Speaker adaptive training: A maximum likelihood approach to speaker normalization. Proceedings of ICASSP, Munich, Germany, 1997: 1043-1046.
  • 4Sinha R and Gales M J F, et al.. The CU-HTK mandaria broadcast news transcription system. Proceedings of ICASSP, Toulouse, France, 2006: 1077-1080.
  • 5Ng Tim, et al.. Progress in the BBN 2007 mandarin speech to text system. Proceedings of ICASSP, Las Vegas, USA, 2008: 1537-1540.
  • 6Su Teng-rong, Wu Ji, and Wang Zuo-ying. Spatial correlation transformation based on minimum covariance. Proceedings of ICASSP, Las Vegas, USA, 2008: 4697-4700.
  • 7吕艳新,孙书学,顾晓辉.基于EMD和能量比的战场声目标分类与识别[J].振动与冲击,2008,27(11):51-55. 被引量:17
  • 8陈湘涛,李明亮,陈玉娟.基于时间序列相似性聚类的应用研究综述[J].计算机工程与设计,2010,31(3):577-581. 被引量:27
  • 9曾番,鹿光,李国宏.基于小波包分析的战场被动声目标特征提取[J].弹箭与制导学报,2010,30(2):240-242. 被引量:1
  • 10祁瑞华,杨德礼,胡润波.基于相关系数加权朴素信念分类模型[J].计算机工程与设计,2010,31(22):4824-4826. 被引量:1

引证文献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