Improved Speech Enhancement Algorithm Based on Bark Bands Noise-Estimation for Non-Stationary Environment

Dong Yu Lu; Guan Yu Tian; Xiao Shan Lu; Xin Ma; Lan Tian

doi:10.4028/www.scientific.net/AMM.385-386.1398

Paper Titles

Energy Based Dual-Microphone Electronic Speech Segregation
p.1381

Emotion Recognition from Speech Signals Using Elicited Data and Fuzzy KDA
p.1385

Voltage Flicker Extraction Based on Wavelet Analysis
p.1389

De-Noise of the High Frequency Corona Current with Wavelet Analysis Method
p.1394

Improved Speech Enhancement Algorithm Based on Bark Bands Noise-Estimation for Non-Stationary Environment
p.1398

Objective Recognition Algorithm of Remote Sensing Image Based on Wavelet Transform
p.1402

An Algorithm of the LMS Adaptive Filtering in Linear Canonical Transform Domain
p.1407

ANFIS Model for Time Series Prediction
p.1411

Recent Frequent Item Mining Algorithm in a Data Stream Based on Flexible Counter Windows
p.1415

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 385-386Improved Speech Enhancement Algorithm Based on...

Improved Speech Enhancement Algorithm Based on Bark Bands Noise-Estimation for Non-Stationary Environment

Abstract:

The conventional spectrum subtraction algorithm cannot effectively suppress the noise under highly non-stationary environment and results in the remaining music noise is often heard in the enhanced speech. In order to improve the speech enhancement performance, a novel denoising algorithm is proposed, which is based on speech endpoint detection using spectrum variance and the dynamic spectrum subtraction in Bark bands. According to human auditory characteristics, the Bark bands spectrums of the noisy speech signal are firstly calculated, and the noise power spectrum of each Bark band is then tracked and estimated by the improved minima controlled recursive averaging method. This noise estimation is adjustable frame by frame and more accurate for non-stationary environment. The experiment results showed that the proposed method can suppress the noise more efficiently than the conventional spectrum subtraction and the remaining music noise is almost eliminated.

You might also be interested in these eBooks

Applied Scientific Research and Engineering Developments for Industry

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 385-386)

Pages:

1398-1401

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.385-386.1398

Citation:

Cite this paper

Online since:

August 2013

Authors:

Dong Yu Lu, Guan Yu Tian, Xiao Shan Lu, Xin Ma, Lan Tian

Keywords:

Bark Bands, Non-Stationary Noise, Spectral Subtraction, Speech Enhancement

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] Thomsa F. Q, Discrete-time speech signal processing. Beijing: Electronics Industry Press, (2004).

Google Scholar

[2] Zhang Rui. Research on Speech Enhancement. Chendu, Sichuan University, (2006).

Google Scholar

[3] Pu Chunjun, Majingxia, Xupeng. Research on endpoint detection method of speech with noise. Compute Application, 2006, 26(11), pp.2685-2690.

Google Scholar

[4] Wang Y., Qu BD, Li JB, An improved endpoint detection algorithm based on band variance. Proceedings of Control and Decision Conference, Wuxi, China, 2007, pp.301-303.

Google Scholar

[5] R. Sundarrajan, C.L. Philips, A noise-estimation algorithm for highly non-stationary environments. Speech Communication, 2006, 48(2), pp.220-231.

DOI: 10.1016/j.specom.2005.08.005

Google Scholar