Noise Estimation Using Mean Square Cross Prediction Error for Speech Enhancement
- 12 July 2010
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 57 (7), 1489-1499
- https://doi.org/10.1109/tcsi.2010.2054930
Abstract
This paper shows the feasibility of noise extraction from noisy speech and presents a two-stage approach for speech enhancement. The preproposed mean square cross prediction error (MSCPE) based blind source extraction algorithm is utilized to extract the additive noise from the noisy speech signal in the first stage. After that a modified spectral subtraction and a modified Wiener filter approach are proposed to extract the speech signal for speech enhancement in the second stage, where all the frequency spectra of the extracted noise are utilized. Theoretical justification shows that the MSCPE-based algorithm can extract desired signal from mixed sources. Experimental results show that the averaged correlation coefficient between the extracted noise and the original additive noise are beyond 85% for Gaussian noise and beyond 75% for real-world noise at SNR = 0 dB, and the proposed speech enhancement approaches perform better than conventional methods, such as spectral subtraction and Wiener filter.Keywords
This publication has 22 references indexed in Scilit:
- Extraction of Desired Signal Based on AR Model with Its Application to Atrial Activity Estimation in Atrial FibrillationEURASIP Journal on Advances in Signal Processing, 2008
- Speech Signal Enhancement Based on MAP Algorithm in the ICA SpaceIEEE Transactions on Signal Processing, 2008
- Speech enhancement employing Laplacian-Gaussian mixtureIEEE Transactions on Speech and Audio Processing, 2005
- Speech Enhancement by MAP Spectral Amplitude Estimation Using a Super-Gaussian Speech ModelEURASIP Journal on Advances in Signal Processing, 2005
- Extension of the signal subspace speech enhancement approach to colored noiseIEEE Signal Processing Letters, 2003
- A multi-band spectral subtraction method for enhancing speech corrupted by colored noisePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Noise power spectral density estimation based on optimal smoothing and minimum statisticsIEEE Transactions on Speech and Audio Processing, 2001
- An adaptive KLT approach for speech enhancementIEEE Transactions on Speech and Audio Processing, 2001
- A statistical model-based voice activity detectionIEEE Signal Processing Letters, 1999
- Enhancement and bandwidth compression of noisy speechProceedings of the IEEE, 1979