Digital Signal Processing

Mohini S. Avatade ¹, Shashikant L. Sahare ¹

Affiliations
1 Cummins College of Engineering for Women, Karve Nagar, University of Pune, IN

Abstract

This paper presents an innovative approach to speech signal de-noising in highly Non-Stationary noise Environment. The proposed technique is based on Independent Component Analysis for Blind Source Separation along with efficient and recent popular Adaptive Wavelet packet Thresholding scheme. An Independent Component Analysis is phenomenon which separates various statistically independent components of input speech signal from an observed mixture vector. An Independent component analysis extracts clean speech signal and noise signal from a mixture of individual sources. Negative Entropy measure of non Gaussianity is used as basic principle of independent component analysis. An Adaptive Wavelet packet Thresholding algorithm is applied to eliminate residual noise which remains in extracted individual signals. Prior to this algorithm, an estimation of noise level is carried out. The Time as well as frequency domain Objective Quality Measures such as Log-Likelihood Ratio (LLR), Segmental Signal to Noise Ratio (SNRseg), Weighted Spectral Slope (WSS), Perceptual Evaluation of speech Quality (PESQ), Itakura-Saito (IS) Ratio are then evaluated for resultant Enhanced speech signal with respect to the original desired signal.

Keywords

Independent Component Analysis, Non Stationary Noise Estimation Technique, Wavelet Packet Transform, Adaptive Thresholding, Objective Quality Measures.

Full Text

     

Sangita K. Bavkar ¹, Shashikant L. Sahare ²

Affiliations
1 PVG’s COET, Pune, IN
2 Cummins College of Engineering for Women, Pune, Maharashtra, IN

Abstract

In this paper we are presenting a speech enhancement algorithm for noisy speech signal using the principal component analysis. Principal component analysis (PCA) is a universal subspace approach which is used for enhancement of speech distorted by the noise. The principal component analysis is based on the eigenvalue analysis; the noisy speech signal eigenvalues are classified into clean speech signal eigenvalues and noisy signal eigenvalues and retaining only clean speech signal eigenvalues estimate enhanced signal. The performance evaluation of PCA based speech enhancement algorithm with different noise estimation methods are analyzed in this paper. Objective and informal listening test shows that proposed method works efficiently with improved minima controlled recursive averaging method. The system performs better noise reduction with negligible residual musical noise when tested with sentences corrupted by noise.

Keywords

Eigenvalue Analysis, Noise Estimation, Noise Variance, Principal Component Analysis (PCA), Speech Enhancement, Subspace Signal.

Full Text