Wavelet-based speech enhancement using time-adapted noise estimation

Sheau Fang Lei, Ying Kai Tung

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Spectral subtraction is commonly used for speech enhancement in a single channel system because of the simplicity of its implementation. However, this algorithm introduces perceptually musical noise while suppressing the background noise. We propose a wavelet-based approach in this paper for suppressing the background noise for speech enhancement in a single channel system. The wavelet packet transform, which emulates the human auditory system, is used to decompose the noisy signal into critical bands. Wavelet thresholding is then temporally adjusted with the noise power by time-adapted noise estimation. The proposed algorithm can efficiently suppress the noise while reducing speech distortion. Experimental results, including several objective measurements, show that the proposed wavelet-based algorithm outperforms spectral subtraction and other wavelet-based denoising approaches for speech enhancement for non-stationary noise environments.

Original languageEnglish
Pages (from-to)2555-2563
Number of pages9
JournalIEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
VolumeE91-A
Issue number9
DOIs
Publication statusPublished - 2008 Sep

Fingerprint

Noise Estimation
Speech Enhancement
Speech enhancement
Wavelets
Subtraction
Wavelet Thresholding
Wavelet Packet Transform
Denoising
Simplicity
Decompose
Experimental Results
Background

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Computer Graphics and Computer-Aided Design
  • Electrical and Electronic Engineering
  • Applied Mathematics

Cite this

@article{4621cd6374e14c17af5cd48c3c969eb5,
title = "Wavelet-based speech enhancement using time-adapted noise estimation",
abstract = "Spectral subtraction is commonly used for speech enhancement in a single channel system because of the simplicity of its implementation. However, this algorithm introduces perceptually musical noise while suppressing the background noise. We propose a wavelet-based approach in this paper for suppressing the background noise for speech enhancement in a single channel system. The wavelet packet transform, which emulates the human auditory system, is used to decompose the noisy signal into critical bands. Wavelet thresholding is then temporally adjusted with the noise power by time-adapted noise estimation. The proposed algorithm can efficiently suppress the noise while reducing speech distortion. Experimental results, including several objective measurements, show that the proposed wavelet-based algorithm outperforms spectral subtraction and other wavelet-based denoising approaches for speech enhancement for non-stationary noise environments.",
author = "Lei, {Sheau Fang} and Tung, {Ying Kai}",
year = "2008",
month = "9",
doi = "10.1093/ietfec/e91-a.9.2555",
language = "English",
volume = "E91-A",
pages = "2555--2563",
journal = "IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences",
issn = "0916-8508",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "9",

}

Wavelet-based speech enhancement using time-adapted noise estimation. / Lei, Sheau Fang; Tung, Ying Kai.

In: IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, Vol. E91-A, No. 9, 09.2008, p. 2555-2563.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Wavelet-based speech enhancement using time-adapted noise estimation

AU - Lei, Sheau Fang

AU - Tung, Ying Kai

PY - 2008/9

Y1 - 2008/9

N2 - Spectral subtraction is commonly used for speech enhancement in a single channel system because of the simplicity of its implementation. However, this algorithm introduces perceptually musical noise while suppressing the background noise. We propose a wavelet-based approach in this paper for suppressing the background noise for speech enhancement in a single channel system. The wavelet packet transform, which emulates the human auditory system, is used to decompose the noisy signal into critical bands. Wavelet thresholding is then temporally adjusted with the noise power by time-adapted noise estimation. The proposed algorithm can efficiently suppress the noise while reducing speech distortion. Experimental results, including several objective measurements, show that the proposed wavelet-based algorithm outperforms spectral subtraction and other wavelet-based denoising approaches for speech enhancement for non-stationary noise environments.

AB - Spectral subtraction is commonly used for speech enhancement in a single channel system because of the simplicity of its implementation. However, this algorithm introduces perceptually musical noise while suppressing the background noise. We propose a wavelet-based approach in this paper for suppressing the background noise for speech enhancement in a single channel system. The wavelet packet transform, which emulates the human auditory system, is used to decompose the noisy signal into critical bands. Wavelet thresholding is then temporally adjusted with the noise power by time-adapted noise estimation. The proposed algorithm can efficiently suppress the noise while reducing speech distortion. Experimental results, including several objective measurements, show that the proposed wavelet-based algorithm outperforms spectral subtraction and other wavelet-based denoising approaches for speech enhancement for non-stationary noise environments.

UR - http://www.scopus.com/inward/record.url?scp=77953470319&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953470319&partnerID=8YFLogxK

U2 - 10.1093/ietfec/e91-a.9.2555

DO - 10.1093/ietfec/e91-a.9.2555

M3 - Article

AN - SCOPUS:77953470319

VL - E91-A

SP - 2555

EP - 2563

JO - IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

JF - IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

SN - 0916-8508

IS - 9

ER -