期刊文献+

基于子带谱熵的二阶CMN语音识别鲁棒性研究

下载PDF
导出
摘要 在自动语音识别系统(ASR)中鲁棒性是一个至关重要的问题,为了抑制训练和测试环境的失配,降低背景噪声和信道传输对语音信号的影响,文章提出了一种基于子带谱熵的二阶CMN语音识别算法。该算法利用子带谱熵在低信噪比下对语音信号进行端点检测具有较高稳健性的特点,将带噪语音分割为背景噪声段和语音信号段,为抑制噪声和信道对语音识别系统的干扰,采用在不同的区间去除各自的倒谱平均值来实现。仿真实验结果表明,该算法克服了传统CMN算法不能处理环境噪声及传输信道对语音信号所产生的非线性畸变的缺点,有效提高了语音识别系统的鲁棒性。
作者 谢杨梅 吕钊
出处 《池州学院学报》 2015年第6期23-26,共4页 Journal of Chizhou University
基金 安徽高校省级优秀青年人才基金项目(2011SQRL162) 池州学院自然科学科研项目(2010ZR010)
  • 相关文献

参考文献8

  • 1Gerkmann T, Krawczyk-Becker M and Le Roux J. Phase pro- cessing for single-channel speech enhancement[J].IEEE Signal Pro- cessing Magazine, 2015, 32 (2): 55-66.
  • 2何勇军,韩纪庆.语音识别中带宽失配的补偿研究[J].计算机学报,2011,34(9):1629-1637. 被引量:4
  • 3夏乐乐,孙永荣,王勇.基于自适应噪声估计的语音增强技术[J].计算机工程与应用,2014,50(23):225-228. 被引量:2
  • 4Atal BS. Effectiveness of linear prediction characteristics of the speech wave for automatic speaker identification and verification[J].J Acoust Soc Am, 1974,55(6):1304-1312.
  • 5Yamamoto K, Jabloun F, einhard K, et al. Robust endpoint de- tection for speeeh recognition based on discriminative feature extraction [C]//IEEE Proeeedings ICASSP, 2006.
  • 6Wu BF and Wang KC. Robust endpoint detection algorithm based on the adaptive band-partitioning spectral entropy in adverse en- vironments[J].IEEE Transaction on speeeh and audio processing, 2005, 13(5): 762-775.
  • 7侯周国,钱盛友,姚畅.短时域语音端点检测中谱熵算法的改进[J].计算机工程与应用,2006,42(21):55-56. 被引量:3
  • 8Misra H, Martigny S, Ikbal S, et al. Multi-resolution spectral entropy feature for robust ASR[C]//IEEE Proceedings ICASSP, 2005.

二级参考文献41

  • 1卜凡亮,王为民,戴启军,陈砚圃.基于噪声被掩蔽概率的优化语音增强方法[J].电子与信息学报,2005,27(5):753-756. 被引量:16
  • 2钱国青,赵鹤鸣.基于改进谱减算法的语音增强新方法[J].计算机工程与应用,2005,41(35):42-43. 被引量:18
  • 3Walker W, Lamere P. Sphinx-4: A flexible open source framework for speech recognition. Sun Microsystenas, 2004: 1-15.
  • 4Li J Y, Deng L, Yu D. A unified framework of HMM adap- tation with joint compensation of additive and convolutive dis- tortions. Computer Speech and Language, 2009, 23: 389- 405.
  • 5Ortega-Garcia J, Gonzalez-Rodriguez L. Overview of speaker enhancement techniques for automatic speaker recognition// Proceedings of the International Conference on Spoken Lan- guage Processing. Philadelpia, 1996:929-932.
  • 6Benesty J, Chen J D, Huang Y T. Microphone Array Signal Processing. Berlin Heidelberg.- Springer-Verlag, 2008.
  • 7You C H, Rahardja S, Member S. Audible noise reduction in eigendomain for speech enhancement. IEEE Transactions on Audio, Speech, and Language Processing, 2007, 15(6): 1753-1765.
  • 8Furui S. Cepstral analysis technique for automatic speaker verification. IEEE Transactions on Acoustics, Speech and Signal Processing, 1981, 29(2): 254-272.
  • 9Barras C, Gauvain J L. Feature and score normalization for speaker verification of cellular data//Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing. Hongkong, 2003, 2: 49-52.
  • 10Pelecanos J, Sridharan S. Feature warping for robust speaker verification//Proceedings of the ISCA Workshop Speaker Recognition. 2001: 213-218.

共引文献6

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