Independent Component Analysis Voice Enhancement  Based on Bessel Function Expansion

doi:10.3969/j.issn.1000-3428.2013.03.062

Computer Engineering ›› 2013, Vol. 39 ›› Issue (3): 311-315. doi: 10.3969/j.issn.1000-3428.2013.03.062

• Networks and Communications • Previous Articles Next Articles

Independent Component Analysis Voice Enhancement Based on Bessel Function Expansion

Independent Component Analysis with Reference(ICA-R) can extract only desired source signal from mixtures of all source signals by incorporating prior information into the learning algorithm as reference signal. The rule of voice signal transmission and Bessel function expansion can describe voice signals. This paper applies Independent Component Analysis with Reference(ICA-R) to extract a target voice signal from mixtures of all source signals by constructing a proper reference signal with Bessel function expansion, computer simulation results and performance analysis demonstrate the method can get better voice enhancement effect under noise interference situation.

(Institute of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650093, China)

Received:2012-03-27 Online:2013-03-15 Published:2013-03-13

基于Bessel函数展开的ICA语音增强

将源信号的先验知识以参考信号的形式引入到独立分量分析(ICA)学习算法中，从混合信号中仅提取期望的源信号。依据语音信号传播机理和Bessel函数展开系数对语音信号的表征能力，给出基于Bessel函数展开的参考信号构建方法，从混合语音信号中提取出期望的语音信号。仿真和性能分析结果表明，该方法能在噪声干扰的情况下达到语音增强的目的。

(昆明理工大学信息工程与自动化学院，昆明 650093)

作者简介:熊志伟(1988－)，男，硕士研究生，主研方向：盲信号处理；全海燕，教授、博士研究生；周荣强，硕士研究生
基金资助:
云南省自然科学基金资助项目(2009ZC048M)

Abstract

Abstract: Independent Component Analysis with Reference(ICA-R) can extract only desired source signal from mixtures of all source signals by incorporating prior information into the learning algorithm as reference signal. The rule of voice signal transmission and Bessel function expansion can describe voice signals. This paper applies Independent Component Analysis with Reference(ICA-R) to extract a target voice signal from mixtures of all source signals by constructing a proper reference signal with Bessel function expansion, computer simulation results and performance analysis demonstrate the method can get better voice enhancement effect under noise interference situation.

Key words: Blind Source Separation(BSS), Independent Component Analysis(ICA), ICA with Reference(ICA-R), Bessel function, Empirical Mode Decomposition(EMD), voice enhancement

摘要： 将源信号的先验知识以参考信号的形式引入到独立分量分析(ICA)学习算法中，从混合信号中仅提取期望的源信号。依据语音信号传播机理和Bessel函数展开系数对语音信号的表征能力，给出基于Bessel函数展开的参考信号构建方法，从混合语音信号中提取出期望的语音信号。仿真和性能分析结果表明，该方法能在噪声干扰的情况下达到语音增强的目的。

关键词: 盲源分离, 独立分量分析, 参考独立分量分析, Bessel函数, 经验模式分解, 语音增强

CLC Number:

TP391

XIONG Zhi-Wei, QUAN Hai-Yan, ZHOU Rong-Jiang. Independent Component Analysis Voice Enhancement Based on Bessel Function Expansion[J]. Computer Engineering, 2013, 39(3): 311-315.

熊志伟, 全海燕, 周荣强. 基于Bessel函数展开的ICA语音增强[J]. 计算机工程, 2013, 39(3): 311-315.

/ / Recommend / Download Citations

URL: http://www.ecice06.com/EN/10.3969/j.issn.1000-3428.2013.03.062

http://www.ecice06.com/EN/Y2013/V39/I3/311

References

[1] 胡航. 语音信号处理[M]. 哈尔滨: 哈尔滨工业大学出版社, 2005.
[2] Comon P. Independent Component Analysis, A New Concept?[J]. Signal Processing, 1994, 36(3): 287-341.
[3] Hyvarinen A, Oja E. A Fast Fixed-point Algorithm for Independent Component Analysis[J]. Neural Computation, 1997, 9(7): 1483-1492.
[4] Zhang Zhilin. Morphological Constrained ICA for Extracting Weak Temporally Correlated Signals[J]. Neuro Computing 2008, 71(7-9): 179-197.
[5] Lu Wei, Rajapakse J C. ICA with Reference[C]//Proc. of the 3rd International Conference on ICA and Blind Source Separation. [S. l.]: Springer, 2001: 120-125.
[6] Lin Qiuhua, Zheng Yongrui, Yin Fuliang, et al. Voice Segregation Using Constrained ICA[C]//Advances in Neural Networks. [S. l.]: Springer, 2004: 755-760.
[7] 林秋华, 郑永端, 殷福亮. 基于参考独立分量分析的语音增强方法[J]. 大连理工大学学报, 2005, 46(6): 915-1000.
[8] 林秋华. 参考独立分量分析算法及应用研究[D]. 大连:大连理工大学, 2005.
[9] Zheng Yongrui, Lin Qiuhua, Yin Fuliang, et al. Voice Enhancement Using ICA with EMD-based Reference ICA[C]//Proc. of the 6th International Conference on Independent Component Analysis and Blind Signal Separation. Berlin, Germany: Springer, 2006: 739-746.
[10] Gopalan K. A Comparison of Speaker Identification Results Using Features Based on Cepstrum and Fourier-Bessel Expansion[J]. IEEE Trans. on Speech and Audio Proceeding, 1999, 7(3): 289-294.
[11] 王浩, 宋巍, 岳振军. 语音信号的傅立叶-贝赛尔展开及其特性初探[C]//全国第一届嵌入式技术联合学术会议论文集. 乌鲁木齐: [出版者不详], 2006.
[12] Chen C S, Gopalan K, Mitra P. Speech Signal Analysis and Synthesis via Fourier-Bessel Representation[C]//Proc. of IEEE International Conference on Acoustics, Speech, and Signal Processing. [S. l.]: IEEE Press, 1985: 497-500.
[13] 王秋生, 段丹辉. 经验模态分解的边界效应处理技术[J].计算机测量与控制, 2006, 14(12): 1671-1673.

[1]	LI Zhenbi,WANG Kang,JIANG Yuanyuan. Variable Step-size EASI Blind Source Separation Algorithm Based on Crosstalk Error and Separation Degree [J]. Computer Engineering, 2017, 43(8): 108-113.
[2]	FU Hui,WU Bin,LI Linfei,ZHANG Hongying. Definition Recovery of Haze Image Based on Dark Channel Prior [J]. Computer Engineering, 2016, 42(7): 232-237.
[3]	ZHANG Fei,GENG Hongqin. Energy Optimization Algorithm in WSN Based on Shadow Fading Channel Modeling [J]. Computer Engineering, 2016, 42(6): 108-112.
[4]	MENG Ming,LU Shaona,MA Yuliang. Electroencephalogram De-noising Method Based on EEMD and Improved Lifting Wavelet [J]. Computer Engineering, 2016, 42(4): 313-317.
[5]	SU Qiao,SHEN Yuehong,XU Pengcheng. Fast Separation Algorithm for Bounded Mixed Signals [J]. Computer Engineering, 2016, 42(2): 86-92.
[6]	FENG Bao,QIN Chuanbo. Blind Signal Separation Algorithm of Brain Image Data Based on Convex Optimization [J]. Computer Engineering, 2015, 41(8): 233-237.
[7]	CHEN Fenglin,LIU Yongbin,FANG Jian,XU Qiang. Equipment Status Feature Extraction Method Based on EMD and JADE [J]. Computer Engineering, 2015, 41(7): 305-309.
[8]	SHAO Jingfeng,WANG Jinfu,BAI Xiaobo,LEI Xia,LIU Congying. Comparison Research on Weaving Data Fitting Method Based on Empirical Mode Decomposition [J]. Computer Engineering, 2015, 41(5): 185-190,196.
[9]	XU Libin,SONG Yuqing,LIU Yi. Restraining Method for End Effect Based on Mirror Extension and Window Function [J]. Computer Engineering, 2015, 41(4): 112-116.
[10]	LIANG Jinping,DONG Weiguang,MAO Xiangde. Study on Fault Feature Extraction and Dimension Reduction Method of Converter [J]. Computer Engineering, 2015, 41(12): 280-287.
[11]	FENG Bao,LIU Xiaogang. Blind Separation Method of Brain Image Data Based on Dictionary Sparsity [J]. Computer Engineering, 2015, 41(12): 269-274.
[12]	WANG Weijia,XIAO Mingqing,ZHANG Lei,CHEN Maocai. Research on Empirical Mode Decomposition Denoising Under Multiwavelet Packet [J]. Computer Engineering, 2015, 41(12): 125-129.
[13]	AI Lingmei,LI Ke,MA Miao. Application of fGn Model in Colon Cancer Gene Expression Dataset Denoising [J]. Computer Engineering, 2015, 41(11): 303-307.
[14]	YAN Wenwu,PAN Feng. Industry Process Fault Classification Based on ICA-ELM [J]. Computer Engineering, 2015, 41(10): 290-294.
[15]	MA Bin, CHEN Jun-jie. Independent Component Analysis Method in Brain Image Data [J]. Computer Engineering, 2014, 40(3): 205-207.

Please choose a citation manager

Content to export

Independent Component Analysis Voice Enhancement Based on Bessel Function Expansion

基于Bessel函数展开的ICA语音增强

PDF

Knowledge

Cited

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics

Comments

模态框（Modal）标题

Please choose a citation manager

Content to export

Independent Component Analysis Voice Enhancement Based on Bessel Function Expansion

基于Bessel函数展开的ICA语音增强

PDF

Knowledge

Cited

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics

Comments