Improving the performance of the minimum statistics noise estimator for single channel speech enhancement

Seung Kyun Ryu, Hong-Goo Kang, Sung Kyo Jung, Dae Hee Youn

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper proposes an algorithm to improve the performance of the noise power spectrum estimation using the minimum statistics (MS). The minimum statistics noise estimator (MSNE) that is most efficient for speech enhancement often underestimates noise power when the signal characteristics changes abruptly. The proposed algorithm improves the accuracy of noise estimation by removing harmonic components of the speech signal. Simulation results verify that the performance of the proposed algorithm is better than that of the conventional algorithm in terms of the segmental SNR (S egS NR) and the spectral distance (SD).

Original languageEnglish
Pages (from-to)582-585
Number of pages4
JournalIEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
VolumeE88-A
Issue number2
DOIs
Publication statusPublished - 2005 Jan 1

Fingerprint

Speech Enhancement
Speech enhancement
Statistics
Estimator
Noise Estimation
Speech Signal
Power spectrum
Power Spectrum
Harmonic
Verify
Simulation

All Science Journal Classification (ASJC) codes

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

Cite this

@article{a10d9b6bc41c4cf58df991af9201ce2b,
title = "Improving the performance of the minimum statistics noise estimator for single channel speech enhancement",
abstract = "This paper proposes an algorithm to improve the performance of the noise power spectrum estimation using the minimum statistics (MS). The minimum statistics noise estimator (MSNE) that is most efficient for speech enhancement often underestimates noise power when the signal characteristics changes abruptly. The proposed algorithm improves the accuracy of noise estimation by removing harmonic components of the speech signal. Simulation results verify that the performance of the proposed algorithm is better than that of the conventional algorithm in terms of the segmental SNR (S egS NR) and the spectral distance (SD).",
author = "Ryu, {Seung Kyun} and Hong-Goo Kang and Jung, {Sung Kyo} and Youn, {Dae Hee}",
year = "2005",
month = "1",
day = "1",
doi = "10.1093/ietfec/E88-A.2.582",
language = "English",
volume = "E88-A",
pages = "582--585",
journal = "IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences",
issn = "0916-8508",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "2",

}

Improving the performance of the minimum statistics noise estimator for single channel speech enhancement. / Ryu, Seung Kyun; Kang, Hong-Goo; Jung, Sung Kyo; Youn, Dae Hee.

In: IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, Vol. E88-A, No. 2, 01.01.2005, p. 582-585.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Improving the performance of the minimum statistics noise estimator for single channel speech enhancement

AU - Ryu, Seung Kyun

AU - Kang, Hong-Goo

AU - Jung, Sung Kyo

AU - Youn, Dae Hee

PY - 2005/1/1

Y1 - 2005/1/1

N2 - This paper proposes an algorithm to improve the performance of the noise power spectrum estimation using the minimum statistics (MS). The minimum statistics noise estimator (MSNE) that is most efficient for speech enhancement often underestimates noise power when the signal characteristics changes abruptly. The proposed algorithm improves the accuracy of noise estimation by removing harmonic components of the speech signal. Simulation results verify that the performance of the proposed algorithm is better than that of the conventional algorithm in terms of the segmental SNR (S egS NR) and the spectral distance (SD).

AB - This paper proposes an algorithm to improve the performance of the noise power spectrum estimation using the minimum statistics (MS). The minimum statistics noise estimator (MSNE) that is most efficient for speech enhancement often underestimates noise power when the signal characteristics changes abruptly. The proposed algorithm improves the accuracy of noise estimation by removing harmonic components of the speech signal. Simulation results verify that the performance of the proposed algorithm is better than that of the conventional algorithm in terms of the segmental SNR (S egS NR) and the spectral distance (SD).

UR - http://www.scopus.com/inward/record.url?scp=24144438687&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=24144438687&partnerID=8YFLogxK

U2 - 10.1093/ietfec/E88-A.2.582

DO - 10.1093/ietfec/E88-A.2.582

M3 - Article

AN - SCOPUS:24144438687

VL - E88-A

SP - 582

EP - 585

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 - 2

ER -