期刊文献+

噪声自适应的多数据流复合子带语音识别方法 被引量:3

Noise Adaptive Multi-stream Hybrid Sub-band Approach for Robust Speech Recognition
在线阅读 下载PDF
导出
摘要 首先针对现有丢失数据语音识别技术中的边缘化(marginalisation)技术在特征运用上的局限,提出了一种倒谱特征分量的可靠性估计方法,将边缘化技术推广到常用的倒谱语音识别系统中;然后利用基于全带和子带倒谱特征的边缘化识别器在不同噪声中的互补性能,提出了一种噪声自适应的多数据流复合子带语音识别方法。实验结果表明,所提识别方法可以自适应地选出全带和子带数据流中受噪声影响较小者并以之为主要依据进行识别,有效地提高了识别系统在多变噪声环境中的鲁棒性。 This paper first proposes a new method for evaluating the reliability of cepstral components and extends the marginalisation technique to cepstral recognizers. Then a noise adaptive multi-stream hybrid sub-band approach is proposed for robust speech recognition by making use of the complemental performances between full-band and sub-band cepstral marginalisation recognizers in different noises. Experimental results show that the proposed approach can turn to the less distorted data stream automatically and improve the robustness of the speech recognizer in various noisy environments effectively.
作者 张军 韦岗
出处 《电子与信息学报》 EI CSCD 北大核心 2006年第7期1183-1187,共5页 Journal of Electronics & Information Technology
基金 国家自然科学青年基金(60502041) 广东省自然科学博士启动基金(65300146)资助课题
关键词 语音识别 丢失数据 边缘化 多数据流 复合子带 Speech recognition, Missing data, Marginalisation, Multi-stream, Hybrid sub-band
  • 相关文献

参考文献9

  • 1Cooke M, Green P, Josifovski L, Vizinho A. Robust automatic speech recognition with missing and unreliable acoustic data.Speech Communication, 2001, 34(3): 267 - 285.
  • 2罗宇,杜利民.基于隐马尔可夫模型局部最优状态路径的数据重建算法[J].电子与信息学报,2004,26(5):722-726. 被引量:8
  • 3PalomAAkim K J, Brown G J, Wang D L. A binaural processor for missing data speech recognition in the presence of noise and small-room reverberation. Speech Communication, 2004, 43(4):361 - 378
  • 4Veth J, Cranen B, Boves L. Acoustic backing-off as an implementation of missing feature theory. Speech Communication, 2001, 34(3): 247-256.
  • 5Okawa S, Bocchieri E, Potamianos A. Multi-band speech recognition in noisy environments. IEEE International Conference on Acoustics, Speech, and Signal Processing,Seattle ,Washington USA, May 12-15, 1998: 641-644.
  • 6Hariharan R, Kiss I, Viikki O. Noise robust speech parameterization using multiresolution feature extraction. IEEE Trans. on Speech and Audio Processing, 2001, 9(8): 856 - 865.
  • 7Huerta J, Stem R. Speech recognition from GSM codec parameters. International Conference on Spoken Language Processsing, Sydney , Australia, November 30-December 4,1998, 4.. 1463- 1466.
  • 8蒋文建,韦岗.噪声下差分复合子带语音识别方法[J].通信学报,2002,23(1):18-24. 被引量:4
  • 9Young S, Kershaw D, Odell J, et al.. The HTK Book (for HTK version 3.1). Cambridge, UK: Cambridge University Tech Services Ltd, 2001.

二级参考文献10

  • 1Morris A C, Cooke M, Green P. Some solutions to the missing feature problem in data classification, with application to noise robust ASR. Proc. ICASSP'98, Seattle, 1998: 737-740.
  • 2Vizinho A, Green P, Cooke M, Josifovski L. Missing data theory, spectral subtraction and signalto-noise estimation for robust ASR: An integrated study. Eurospeech'99, Budapest, 1999, vol.5:2407-2410.
  • 3Cooke M, Green P, Josifovski L, Vizinho A. Robust automatic speech recognition with missing and unreliable acoustic data. Speech Communication, 2001, 34(3): 267-285.
  • 4Raj B, Seltzer M, Stern R. Reconstruction of damaged spectrographic features for robust speech recognition. In Proceedings ICSLP'00, Beijing, China, October 2000, vol.1: 375-360.
  • 5Raj B. Reconstruction of incomplete spectrograms for robust speech recognition. [Ph.D Thesis],ECE Department, Carnegie Mellon University, April, 2000.
  • 6Rabiner L R, Juang B H. Fundamentals of Speech Recognition. Prentice-Hall Press, 1993, Ch.6:321-389.
  • 7Steve Young, Dan Kershaw, Julian Odell, Dave Ollason, Valtcho Valtchev, Phil Woodland. The HTK Book (for HTK Version 3.0), Microsoft Corporation, 2000.
  • 8韦晓东,朱杰,胡光锐.汽车噪声中自动语音的识别技术[J].上海交通大学学报,1998,32(10):10-13. 被引量:6
  • 9刘加.汉语大词汇量连续语音识别系统研究进展[J].电子学报,2000,28(1):85-91. 被引量:51
  • 10蒋文建,韦岗.基于掩蔽特性的噪声环境下语音识别新特征[J].声学学报,2001,26(6):516-520. 被引量:10

共引文献10

同被引文献24

引证文献3

二级引证文献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