期刊文献+

基于缺失数据补偿的鲁棒语音识别

Robust Speech Recognition Based on Missing-Data Imputation
在线阅读 下载PDF
导出
摘要 针对实际环境中语音信号的时频分量普遍存在部分缺失或严重失真的问题,在已知语音先验知识的条件下,提出了一种利用可靠时频分量对缺失数据进行补偿的方法。利用贝叶斯准则,将最优补偿转化为求解后验概率最大化的问题,并利用缺失数据自身的能量信息,给出了一种局部最优补偿的方法。实验表明,该方法在各种噪声、信噪比环境下,综合性能优于传统的鲁棒语音识别技术;采用缺失信息对补偿进行限定,在低信噪比下鲁棒性能有了明显的提高。 Data missing is a natural occurrence in the real environment. According to the prior speech distribution, a missing data imputation method is proposed using the reliable data compo-nent. By the Bayesian rule, solving optimal imputation comes down to finding the value which maxi-mizes the posterior probability, and a suboptimal data imputation is proposed according to energy of the missing data. The results of the experiment show that the proposed method outperforms the state-of-the-art robust speech in different SNR environments; especially when the SNR is low, the energy bounded imputation exhibits an obvious improvement in robustness.
出处 《信息工程大学学报》 2012年第4期411-415,共5页 Journal of Information Engineering University
基金 国家自然科学基金资助项目(61175017)
关键词 缺失数据补偿 鲁棒语音识别 贝叶斯准则 missing-data imputation robust speech recognition Bayesian rule
作者简介 牛铜(1983-),男,博士生,主要研究方向为噪声环境下的语音识别、语音增强 李弼程(1970-),男,教授,博士生导师,主要研究方向为智能信息处理
  • 相关文献

参考文献10

  • 1吕勇,吴镇扬.基于隐马尔可夫模型与并行模型组合的特征补偿算法[J].东南大学学报(自然科学版),2009,39(5):889-893. 被引量:4
  • 2Rogier C, van Dalen, Mark J, et al. Extended VTS for Noise-Robust speech Recognition [ Jl. IEEE Tran. On Audio, Speech, and Language Processing, 2011 , 19 (4) : 733-742.
  • 3Araki S, Makino S, Sawada H, et al. Underdetermined blind separation of convolutive mixtures of speech with directivity pattern based mask and ica[ C] J/Proc. 5th Int. Conf. Independent Compon. 2004: 898-905.
  • 4Hu G, Wang D L. Monaural speech segregation based on pitch tracking and amplitude modulation[ J]. IEEE Trans. Neural Netw, 2004, 15(5): 1135-1150.
  • 5Roman N, Wang D L, Brown G J. Speech segregation based on sound localization[ J]. Acoust. Soc. Amer, 2003, 114: 2236 -2252.
  • 6Cooke M P, Morris A Green. Missing data techniques for robust speech recognition[ C ]//Proceeding of the international con- ference on acoustics, speech and signal processing. 1997: 863-866.
  • 7Martin Cooke, Phil Green, Ljubomir Josifovski, et al. Robust automatic speech recohnition with missing and unreliable a- coustic data[J]. Speech Communication, 2001, 34: 267-285.
  • 8Yu Luo, Limin Du. Single gauss model set-based data imputation method for complex ASR task [ C ]//ICASSP. 2003: 564- 567.
  • 9Soundararaian Srinivasan, DeLiang Wang. Transforming binary uncertainties for robust speech recognition[ J]. IEEE Trans. On Audio, 2007, 15(7): 2130-2140.
  • 10理查德.使用多元统计分析[M].6版.陆璇,译.北京:清华大学出版社,2008:123-140.

二级参考文献11

  • 1孙暐,吴镇扬.基于独立感知理论的鲁棒语音识别算法[J].东南大学学报(自然科学版),2005,35(4):506-509. 被引量:2
  • 2赵蕤,王作英.语音识别中信道和噪音的联合补偿[J].声学学报,2006,31(5):466-470. 被引量:11
  • 3Nasersharif B, Akbari A. SNR-dependent compression of enhanced Mel sub-band energies for compensation of noise effects on MFCC features [J ]. Pattern Recognition Letters, 2007,28( 11 ) : 1320 - 1326.
  • 4Cui X, Alwan A. Noise robust speech recognition using feature compensation based on polynomial regression of utterance SNR [ J ]. IEEE Transactions on Speech and Audio Processing, 2005, 13(6) : 1161 -1172.
  • 5Barreaud V, Illina I, Fohr D. On-line stochastic matching compensation for non-stationary noise [ J ]. Computer Speech and Language, 2008, 22 ( 3 ) : 207 - 229.
  • 6Moreno P J. Speech recognition in noisy environments [ D]. Pittsburgh, Pennsylvania, USA: Carnegie Mellon University, 1996: 79 - 126.
  • 7Kim W, Kwon O, Ko H. PCMM-based feature compensation schemes using model interpolation and mixture sharing [ C ]//IEEE International Conference on Acoustics, Speech, and Signal Processing. Montreal, Canada, 2004:989-992.
  • 8Kim W, Hansen J H L. Feature compensation in the cepstral domain employing model combination [ J ]. Speech Communication, 2009, 51 (2) : 83 - 96.
  • 9Sasou A, Asano F, Nakamura S, et al. HMM-based noise-robust feature compensation [ J]. Speech Communication, 2006, 48 (9) : 1100 - 1111.
  • 10Gales M J F, Young S J. Robust speech recognition in additive and convolutional noise using parallel model combination [ J ]. Computer Speech and Language, 1995, 9(4): 289-307.

共引文献3

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