A New Algorithm Simulation Study of Wavelet Package Speech De-noising

Chapter
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 128)

Abstract

When input signal has low SNR, the commonly used wavelet de-noising algorithm will cause envelope distortion problem for reconstructed signal. In order to overcome this, this paper presents a new algorithm of wavelet pocket adaptive threshold de-noising. This new wavelet threshold algorithm is obtained based on sub-band signal to noise energy ratio. It can fit the human auditory characteristics, closely tracking the energy changes of s speech signal and accurately identify the formants. Through formant relevant sub-band data further detailed treatment to avoid false positives distortion. The simulation results show that the proposed algorithm is more effective than traditional algorithm for low signal-to-noise ratio input. It either removes noise as much as possible to improve the output SNR, or effectively reduces signal reconstruction distortion. When this new algorithm is combined with energy spectral subtraction, it can further improve the quality of speech de-noising.

Keywords

speech formant auditory perception adaptive threshold wavelet package de-noising 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Wang, W., Yang, D.-C.: Speech Enhancement Using Wavelet Packet Transfom Based on Auditory Model. Journal of Nanjing University Natural Sciences 37(5), 630–636 (2001)Google Scholar
  2. 2.
    Tian, Y.-J., Zuo, H.-W., Dong, Y.-M.: A new algorithm of wavelet package denoising based on Bark adaptive node threshold. Applied Acoustics 30(1) (2011)Google Scholar
  3. 3.
    Yang, L., Yang, H.-B., Gao, X.-C.: Speech Enhancement Algorithm Based on Tracing the Formant. Journal of Electronics & Information Technology 31(10), 2536–2540 (2009)MathSciNetGoogle Scholar
  4. 4.
    Ackenhusen, J.G.: Real-Time signal processing: Design and implementation of signal processing systems, vol. 06, pp. 173–342. Pearson Education Inc. Publishing as prentice Hall PTR (2006)Google Scholar
  5. 5.
    Weickert, T., Kiencke, U.: Adaptive Estimation of Periodic Noise Energy Distributions for Speech Enhancement. In: 9th IFAC Workshop ALC0SP 2007 (2007)Google Scholar
  6. 6.
    Han, J.-Q., Zhang, L., Zhen, T.-R.: Speech signal processing, pp. 72–74. Publishing house of Qinghua University, Beijing (2004)Google Scholar
  7. 7.
    Gao, Y.-Z., Li, Y.-M., Xu, D.-M.: Wavelet Shrinkage Parameters Selection in Speech Enhancement. Journal of Data Acquisition & Processing 24(31), 290–294 (2009)Google Scholar
  8. 8.
    Sun, X., Zou, X.: Objective Speech Quality Evaluation Based on Weighted Bark Spectral Distortion. Journal of Data Acquisition & Processing 21(3), 302–306 (2006)MathSciNetGoogle Scholar

Copyright information

© Springer-Verlag GmbH Berlin Heidelberg 2012

Authors and Affiliations

  1. 1.Modern Education and Technology CenterTechnology University of QingdaoQingdaoChina
  2. 2.School of Civil EngineeringTechnology University of QingdaoQingdaoChina
  3. 3.Technology University of QingdaoQingdaoChina

Personalised recommendations