Research on the Speech Enhancement Method based on PCA/KLT Algorithms

Guang Yan Wang; Yan Xiang Geng; Xiao Qun Zhao

doi:10.4028/www.scientific.net/AMM.239-240.1274

Paper Titles

A Simplified Carrier Frequency Offset Estimation Algorithm for Burst Digital Transmission
p.1255

The Characteristic Parameters Extraction of the Drawing Forming Cracks Based on Time-Series Analysis and ICA
p.1259

An Improved Speech Blind Separation Algorithm Based on Non-Linear Function
p.1264

Applications of Statistical Directional Characteristics in Synthetic Aperture Radar Imagery Interpretation
p.1269

Research on the Speech Enhancement Method based on PCA/KLT Algorithms
p.1274

Bridging a Gap in Independent Component Analysis
p.1279

A QRS Wave Detection Algorithm Based on Complex Wavelet Transform
p.1284

A Particle Swarm Optimization with Variety Factor
p.1291

Projection Position-Based Sequential Pattern Mining Algorithm
p.1298

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 239-240Research on the Speech Enhancement Method based on...

Research on the Speech Enhancement Method based on PCA/KLT Algorithms

Abstract:

In this paper, we propose a speech enhancement technique in terms of subspace methods to reduce the white or colored noise in strong background noise environment. This subspace approach based on Karhunen-Loève transform (KLT) and implemented via Principal Component Analysis (PCA). The subspace selection provided by the minimum description length (MDL) criterion. An offset factor generated from the white noise was used to modify the variance to adapt to the specified colored noise. The objective speech quality measures SegSNR have been introduced to evaluate the performance of the proposed method in time domain. A large amount of data and figures testify that our algorithm provides high performance for a large scale of input signal-to-noise ratio (-5~10dB). The performance of our algorithm is assessed in white and colored noise.

You might also be interested in these eBooks

Measurement Technology and its Application

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 239-240)

Pages:

1274-1278

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.239-240.1274

Citation:

Cite this paper

Online since:

December 2012

Authors:

Guang Yan Wang, Yan Xiang Geng, Xiao Qun Zhao

Keywords:

Colored Noise, Karhunen-Loève Transform (KLT), Minimum Description Length (MDL) Criterion, Principal Component Analysis (PCA), Speech Enhancement

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] M. Dendrinos, S. Bakamidis, G. Carayannis, Speech Communication, Vol. 10, No. 1, (1991), pp.45-57.

DOI: 10.1016/0167-6393(91)90027-q

Google Scholar

[2] Y. Ephraim, H. L. Van Trees, IEEE Transactions on Speech and Audio Processing , Vol. 3, No. 4, (1995), pp.251-266.

Google Scholar

[3] A. Rezayee, S. Gazor, IEEE Transactions on Speech and Audio Processing, Vol. 9, No. 2, (2001), pp.87-95.

DOI: 10.1109/89.902276

Google Scholar

[4] H. Lev-Ari, Y. Ephraim, IEEE Signal Processing Letters , Vol. 10, No. 4, (2003), pp.104-106.

Google Scholar

[5] H. W. Lou, G. R. Hu, Journal of Sound and Vibration , Vol. 268, No. 4, (2003), pp.717-729.

Google Scholar

[6] J. Hansen, B. Pellom, in ICSLP-98: International Conference on Spoken Language Processing. (ISCA, Sydney, Australia, 1998), vol. 7, pp.2819-2822.

Google Scholar

[7] P. Scalart, L. JV CNET Filho, in IEEE International Conference on Acoustics, Speech and Signal Processing, 1996. ICASSP 1996. (Atlanta, GA, 1996), Vol. 2, pp.629-632.

DOI: 10.1109/icassp.1996.543199

Google Scholar