一种改进的奇异值分解语音增强方法

张丽艳; 殷福亮

doi:10.3724/SP.J.1146.2005.01655

一种改进的奇异值分解语音增强方法

doi: 10.3724/SP.J.1146.2005.01655

张丽艳^{①②;殷福亮},
殷福亮

基金项目:

国家自然科学基金(60372082，60172073)和教育部跨世纪优秀人才基金资助课题

计量
- 文章访问数: 3515
- HTML全文浏览量: 80
- PDF下载量: 1045
- 被引次数: 0
出版历程
- 收稿日期: 2006-12-22
- 修回日期: 2007-06-22
- 刊出日期: 2008-02-19

An Improved Speech Enhancement Method Based on SVD

Zhang Li-yan^{①②;殷福亮},
Yin Fu-liang

摘要

摘要: 该文将多麦克语音增强方法用于单麦克语音增强，给出了一种改进的奇异值分解语音增强方法。该方法首先对输入矩阵进行雅克比奇异值分解，用得到的奇异值矢量构造语音增强滤波器；然后用输入矩阵与滤波器权矢量相乘来构造各路信号；最后采用麦克风阵列波束形成的方法，得到增强后的语音信号。仿真结果表明，该方法能有效地去除加性噪声，并且改善了语音质量。
- 语音增强; 雅克比奇异值分解; 波束形成
Abstract: This paper applies the microphone array speech enhancement method to the single microphone speech enhancement method, and proposes a new speech enhancement method based on Singular Value Decomposition(SVD). First, for the input matrix, the method adopts Jacobi singular value decomposition to get the singular value vector as the speech enhancement filter. Then the method multiplies the input matrix and the coefficient matrix of the filter to constitute the so called each channel signals. Finally, the method adopts microphone array beamforming method to gain the enhanced speech. The simulation shows that the proposed method can get rid of the addictive noise very well, and improve the speech quality.

HTML全文

参考文献(1)

Boll S F. Suppression of acoustic noise in speech usingspectral subtraction[J].IEEE Trans. on Acoustics, Speech, andSignal Processing.1979, 27(2):113-120[2]Flanagan J L, Johaston J D, and Zahn T, et al..Computer-steered microphone arrays for sound transductionin large rooms[J].Journal of Acoustical Society of America.1985,78(5):1508-1518[3]Griffiths L J and Jim C W. An alternative approach tolinearly constrained adaptive beamforming[J].IEEE Trans. onAntennas and Propagation.1982, 30(1):27-34[4]Veen A J D, Deprettere E F, and Swindlhurst A L.Subspace-based signal analysis using singular valuedecomposition[J].Proc. IEEE.1993, 81(9):1277-1308[5]Doclo S and Moonen M. GSVD-based optimal filtering forsingle and multimicrophone speech enhancement. IEEETrans. on Signal Processing, 2002, 50(9): 2231-2244.[6]Doclo S and Moonen M. Robustness analysis of GSVD basedoptimal filtering and generalized sidelobe canceller forhearing aid applications. Proc. of the 1999 IEEEInternational Workshop on Acoustic Echo and Noise Control(IWAENC99), Pocono manor, Pennsy Lvania, USA, Sept.1999: 80-83.[7]Zhang Liyan.[J].Yin Fuliang, and Hou Daiwen. A SVD-basedbeamforming speech enhancement algorithm. Proc. of theInternational Conference on Electronic Measurement Instruments (ICEMI2005), Beijing, China.2005,:-[8]Luk FT. Computing the singular-value decomposition on theILLIAC IV[J].ACM Trans. on Mathematical Software.1980,6(4):524-539[9]Comon P and Golub G H. Tracking a few extreme singularvalues and vectors in signal processing[J].Proc. IEEE.1990,78(8):1327-1343[10]戈卢布 G H，范洛恩 C F，著，袁亚湘，等译. 矩阵计算. 北京：科学出版社，2001: 77-78.Golub G H and Vanloan C F. Matrix Computation. Beijing:Science Press, 2001: 77-78.[11]周长发. 科学与工程数值方法. 北京：清华大学出版社，2002:119-129.Zhou Chang-fa. Science and Engineering NumericalComputation Method. Beijing: Tsighua University Press,2002: 119-129.[12]张贤达. 矩阵分析与应用. 北京：清华大学出版社，2004:367-380.Zhang Xian-da. Matrix Analysis and Applications. Beijing:Tsighua University Press, 2004: 367-380.[13]张贤达．信号处理中的线性代数. 北京：科学出版社，1997:322-334.Zhang Xian-da. Linear Algebra Applied in Signal Processing,Beijing: Science Press, 2001: 322-334.[14]易克初，田斌，付强. 语音信号处理. 北京：国防工业出版社，2000: 97-100.Yi Ke-chu, Tian Bin, and Fu Qiang. Speech Signal Processing.Beijing: National Defense Industry Process, 2000: 97-100.[15]Fischer S and Simmer K U. Beamforming microphone arraysfor speech acquisition in noisy environments. SpeechCommunication, 1996, 30(3): 215-227.

施引文献

资源附件(0)

访问统计