融合统计模型与EMD的宽带话音增强方法

doi:10.3969/j.issn.1000-436x.2013.08.013

Abstract

Abstract:

A combined wideband speech enhancement method based on statistical model and empirical mode decomposition (EMD) was proposed.First,statistical model was used to eliminate the main noise component in noisy speech.Then,the residual noise was further suppressed by a post-processing module which is a speech enhancement algorithm with voice activity detection (VAD) based on EMD.The advantages of the two methods were combined effectively.The performance of the proposed method was evaluated under the standard of ITU-T G160.The experimental results indicate that the algorithm is more effective for improving the SNR in the different noise environments than classical statistical model approach.Meanwhile,in low SNR conditions,musical noise is reduced effectively,and the speech sounds more comfortable.

Key words: wideband speech enhancement, statistical model, EMD, VAD

Xuan ZHOU,Chang-chun BAO,Bing-yin XIA. Combined wideband speech enhancement method based on statistical model and EMD[J]. Journal on Communications, 2013, 34(8): 95-101.

Figures/Tables 10

References 14

[1]	LIM J S , OPPENHEIM A V . Enhancement and bandwidth compression of noisy speech[J]. Proceedings of the IEEE, 1979,67(12): 1586-1604.
[2]	BOLL S F . Suppression of acoustic noise in speech using spectral subtraction[J]. IEEE Trans Acoust Speech Signal Process, 1979,27(2): 113-120.
[3]	JENSEN S H , HANSEN P C , HANSEN S D ,et al. Reduction of broadband noise in speech by truncated QSVD[J]. IEEE Trans Speech and Audio Process, 1995,3(6): 439-448.
[4]	DONOHO D L . De-noising by soft-thresholding[J]. IEEE Trans on Information Theory, 1995,41(3): 613-627.
[5]	EPHRAIM Y , MALAH D . Speech enhancement using a minimum mean-square error short-time spectral amplitude estimator[J]. IEEE Trans Acoust Speech Signal Process, 1984,32(6): 1109-1121.
[6]	BOUDRAA O , CEXUS J C , SAIDI Z . EMD-based signal noise reduction[J]. Int J Sig Process, 2004(1): 33-37.
[7]	ESCH T , HEESE F , GEISER B et al ,et al. Wideband noise suppression supported by artificial bandwidth extension techniques[A]. Proc IEEE Int Conf on Acoustics,Speech,and Signal Processing[C]. Dallas,USA, 2010. 4790-4793.
[8]	ITU-T G160.Voice Enhancement Devices[S]. 2008.
[9]	LOTTER T , VARY P . Speech enhancement by MAP spectral amplitude estimation using a super-Gaussian speech model[J]. EURASIP J Appl Signal Process, 2005(7): 1110-1126.
[10]	LOIZOU P . Speech enhancement based on perceptually motivated Bayesian estimators of the speech magnitude spectrum[J]. IEEE Trans Speech Audio Process, 2005,13(5): 857-869.
[11]	HUANG N E , SHEN Z , LONG S R ,et al. The empirical mode decomposition and the Hilbert spectrum for non-linear and non-stationary time series analysis[A]. Proc Royal Soc London[C]. London,UK, 1998. 903-995.
[12]	LI X Y , ZOU X J , ZHANG R B ,et al. Method of speech enhancement based on Hilbert-huang transform[A]. Proceedings of the 7th World Congress on Intelligent Control and Automation[C]. Chongqing,China, 2008. 8414-8418.
[13]	COHEN I , BERDUGO B . Noise estimation by minima controlled recursive averaging for robust speech enhancement[J]. IEEE Signal Process Lett, 2002,9(1): 12-15.
[14]	ITU-T P862.Perceptual Evaluation of Speech Quality (PESQ):An Objective Method for End-to-end Speech Quality Assessment of Narrow-band Telephone Networks and Speech Codecs[S]. 2001.

Metrics

Recommended 0

No Suggested Reading articles found!

测试项目指标		参考方法	本文方法
期望噪声衰减/dB	Q_m	21.49	23.24
实际噪声/dB	Q_n1	21.57	23.24
衰减和话音/dB	Q_n2	21.62	23.25
电平衰减/dB	Q_s	0.54	0.54
	T₁	0.70	0.70
收敛时间/s	T₂	0.29	0.29
	T₃	0.01	0.01

测试项目指标		参考方法	本文方法
	SNRI/dB	12.13	12.93
G160	TNLR/dB	?19.90	?20.74
	DSN/dB	?1.514	?1.528
	6 dB	0.291 3	0.324 1
PESQI	12 dB	0.250 8	0.301 9
	18 dB	0.211 7	0.258 4

Combined wideband speech enhancement method based on statistical model and EMD

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 14

Related Articles 10

Metrics

Recommended 0

[1]	Yuntian FENG, Xia WU, Xiong XU, Rongqing ZHANG. Research on ionospheric parameters prediction based on deep learning [J]. Journal on Communications, 2021, 42(4): 202-206.
[2]	Limin MA,Jiahui WANG. Invertible secret image sharing scheme based on improved FEMD [J]. Journal on Communications, 2019, 40(7): 48-56.
[3]	Da-wei ZHANG,Chang-chun BAO,Bing-yin XIA. Source localization based on time delay estimation in complex environment [J]. Journal on Communications, 2014, 35(1): 183-190.
[4]	. Combined wideband speech enhancement method based on statistical model and EMD [J]. Journal on Communications, 2013, 34(8): 13-101.
[5]	. Fuzzy adaptive algorithm based on modified current statistical model for vehicle positioning [J]. Journal on Communications, 2013, 34(7): 21-190.
[6]	Zhen-hong SHAO,Wen-feng3 LI,Yi WU,Qiong YANG,Lian-feng SHEN. Fuzzy adaptive algorithm based on modified current statistical model for vehicle positioning [J]. Journal on Communications, 2013, 34(7): 184-190.
[7]	Wai-xi LIU,Shun-zheng YU,Ying GAO,Xiao HU. Opportunistic network coding based on prediction [J]. Journal on Communications, 2013, 34(4): 36-46.
[8]	. Opportunistic Network coding based on prediction [J]. Journal on Communications, 2013, 34(4): 5-46.
[9]	Bo GAO,Qin-yu ZHANG,Yong-sheng LIANG,Ning-ning LIU,Cheng-bo HUANG,Nai-tong ZHANG. Predicting self-similar networking traffic based on EMD and ARMA [J]. Journal on Communications, 2011, 32(4): 47-56.
[10]	Xu-chong LIU,Yong LUO,Jian-xin WANG,Jie WANG. Watermarking algorithm for image authentication based on second generation Bandelet [J]. Journal on Communications, 2010, 31(12): 123-130.