期刊文献+
任意字段
题名或关键词
题名
关键词
文摘
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息

基于概率加权平均的Mel子带特征重建算法 被引量：1

Probability-Weighted Average Algorithm for Mel-Frequency Filter-Bank Vector Reconstruction
在线阅读 下载PDF
导出
摘要 本文提出基于概率加权平均的Mel子带特征数据重建算法 .该算法选择K个最优重建结果的概率加权平均作为被加性噪声掩蔽的语音特征分量的估计 .实验结果表明 ,基于概率加权平均的语音特征数据重建算法降低了重建误差 ,减少了帧间突变现象 ,增强了Mel子带特征的帧间连续性 ,从而显著提高了语音识别系统对加性噪声的鲁棒性能 . Probability weighted average (PWA) algorithm is proposed to reconstruct Mel-frequency filter-bank vectors. The probability-weighted average of K-best reconstructed missing components of Mel-frequency filter-bank vectors is taken as the estimation of components masked by additive noise. Experimental results show that PWA algorithm can reduce reconstruction error, increase the continuity between neighbor Mel-filterbank vectors and greatly improve automatic speech recognition (ASH) system's robustness against additive noise.
作者 罗宇 杜利民
机构地区 中科院声学所语音交互技术研究中心
出处 《电子学报》 EI CAS CSCD 北大核心 2004年第10期1738-1741,共4页 Acta Electronica Sinica
基金 国家 973重点基础研究发展项目 (No .G1 9980 30 50 5)
关键词 子带 加性噪声 语音特征 数据重建 语音识别系统 加权平均 掩蔽 概率 重建算法 分量 缺失特征方法 Algorithms Data processing Estimation Probability Spurious signal noise Vectors
分类号 TN912.34 [电子电信—通信与信息系统] TN919.3 [电子电信—通信与信息系统]
  • 相关文献

参考文献11

  • 1A Vizinho,P Green,M Cooke,L Josifovski.Missing data theory,spectral subtraction and signal-to-noise estimation for robust ASR:An integrated study[A].Eurospeech'99[C].Budapest,1999.
  • 2Martin Cooke,Phil Green,Ljubomir Josifovski and Ascension Vizinho.Robust ASR with unreliable data and minimal assumptions[A].Robust 99[C].Tamper,Finland.
  • 3Morris A C,Cooke M,Green P.Some solutions to the missing feature problem in data classification,with application to noise robust ASR[A].Proc.ICASSP'98[C].1998.737-740.
  • 4Ljubomir Josifovski,Martin Cooke,Phil Green,Ascension Vizinho.State based imputation of missing data for robust speech recognition and speech enhancement[A].in Eurospeech[C].1999.6.2833-2836.
  • 5Jon Barker,Ljubomir Josifovski,Martin Cooke and Phil Green.Soft decisions in missing data techniques for robust automatic speech recognition[A].ICSLP-2000,Beijing[C].2000.373-376.
  • 6B Raj,M L Seltzer,R M Stern.Reconstruction of damaged spectrographic features for robust speech recognition[A].Proceedings of the International Conference on Spoken Language Processing[C].October,2000,Beijing,China.
  • 7M L Seltzer,B Raj,R M Stern.Classifier-based mask estimation for missing feature methods of robust speech recognition[A].Proceedings of the International Conference on Spoken Language Processing[C].October,2000,Beijing,China.
  • 8Philippe Renevey,Rolf Vetter,Jens Kraus.Robust speech recognition using missing feature theory and vector quantization[A].Eurospeech 2001[C].Scandinavia,2001.1107.
  • 9Lippmann R P,Carlson B A.Using missing feature theory to actively select features for robust speech recognition with interruptions,filtering and noise[A].Proc Eurospeech'97 [C].Rhodes,Greece,September 1997.KN37-40.
  • 10Steve Young,Dan Kershaw,Julian Odell,Dave Ollason,Valtcho Valtchev,Phil Woodland,The HTK Book ( for HTK Version 3.1) [M].

同被引文献18

  • 1李凌,曾以成,雷雄国.EMD在说话人辨认中的应用[J].湘潭大学自然科学学报,2006,28(3):108-111. 被引量：6
  • 2QIAN S, MORRIS J M. Wigner distribution decomposition and cross-terms delete representation[J]. Signal Processing, 1992,27 : 125 - 144.
  • 3WONG K M, JIN Q. Estimation of the time-varying frequency of a signal: The cramer-ral bound and the ap- plication of Wigner distribution[J]. IEEE Trans Signal Processing, 1990,1 770:358-375.
  • 4HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-station time series analysisI-J3. The Royal Society, 1998, A454 : 903- 995.
  • 5HUANG N E,WU M L, QU W, et al. Applications of Hilbert-Huang transform to nonstationary financial time series analysis[J]. Applied Stochastic Models in Business and Industry,2003,19:245-268.
  • 6YANG Z H, QI D X,YANG L H. A novel approach for detecting pitch based on Hilbert-Huang Transform [R.. Sun Yat-sen University, 2004.
  • 7YAN R Q, GAO R X. A tour of the Hilbert-Huang transform: an empirical tool for signal analysis[J3. IEEE Instrumentation . Measurement Magazine, 2007,10 (5) : 40 - 45.
  • 8陈迪,龚卫国,李波.噪声鲁棒性说话人识别语音高频加权MFCC提取[J].仪器仪表学报,2008,29(3):668-672. 被引量：15
  • 9吴小羊,刘天佑.基于时频重排的地震信号Wigner-Ville分布时频分析[J].石油地球物理勘探,2009,44(2):201-205. 被引量：21
  • 10刘亚丽,杨鸿武,黄德智.基于加权Mel倒谱系数的说话人识别[J].计算机应用与软件,2009,26(9):24-27. 被引量：3

引证文献1

二级引证文献2

电子学报
2004年 第10期

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