Impact and High-Pitch Noise Suppression Based on Spectral Entropy

Arata KAWAMURA; Noboru HAYASAKA; Naoto SASAOKA

doi:10.1587/transfun.E99.A.777

Impact and High-Pitch Noise Suppression Based on Spectral Entropy

Arata KAWAMURA, Noboru HAYASAKA, Naoto SASAOKA

Full Text Views

0

Share
Cite this

Summary :

We propose an impact and high-pitch noise-suppression method based on spectral entropy. Spectral entropy takes a large value for flat spectral amplitude and a small value for spectra with several lines. We model the impact noise as a flat spectral signal and its damped oscillation as a high-pitch periodic signal consisting of spectra with several lines. We discriminate between the current noise situations by using spectral entropy and adaptively change the noise-suppression parameters used in a zero phase-based impact-noise-suppression method. Simulation results show that the proposed method can improve the perceptual evaluation of the speech quality and speech-recognition rate compared to conventional methods.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E99-A No.4 pp.777-787

Publication Date: 2016/04/01

Publicized

Online ISSN: 1745-1337

DOI: 10.1587/transfun.E99.A.777

Type of Manuscript: PAPER

Category: Engineering Acoustics

Authors

Arata KAWAMURA
  Osaka University
Noboru HAYASAKA
  Osaka Electro-Communication University
Naoto SASAOKA
  Tottori University

Keyword

zero phase signal, speech enhancement, noise suppression, impact noise, damped oscillation

Cite this

Copy

Arata KAWAMURA, Noboru HAYASAKA, Naoto SASAOKA, "Impact and High-Pitch Noise Suppression Based on Spectral Entropy" in IEICE TRANSACTIONS on Fundamentals, vol. E99-A, no. 4, pp. 777-787, April 2016, doi: 10.1587/transfun.E99.A.777.
Abstract: We propose an impact and high-pitch noise-suppression method based on spectral entropy. Spectral entropy takes a large value for flat spectral amplitude and a small value for spectra with several lines. We model the impact noise as a flat spectral signal and its damped oscillation as a high-pitch periodic signal consisting of spectra with several lines. We discriminate between the current noise situations by using spectral entropy and adaptively change the noise-suppression parameters used in a zero phase-based impact-noise-suppression method. Simulation results show that the proposed method can improve the perceptual evaluation of the speech quality and speech-recognition rate compared to conventional methods.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1587/transfun.E99.A.777/_p

Copy

@ARTICLE{e99-a_4_777,
author={Arata KAWAMURA, Noboru HAYASAKA, Naoto SASAOKA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Impact and High-Pitch Noise Suppression Based on Spectral Entropy},
year={2016},
volume={E99-A},
number={4},
pages={777-787},
abstract={We propose an impact and high-pitch noise-suppression method based on spectral entropy. Spectral entropy takes a large value for flat spectral amplitude and a small value for spectra with several lines. We model the impact noise as a flat spectral signal and its damped oscillation as a high-pitch periodic signal consisting of spectra with several lines. We discriminate between the current noise situations by using spectral entropy and adaptively change the noise-suppression parameters used in a zero phase-based impact-noise-suppression method. Simulation results show that the proposed method can improve the perceptual evaluation of the speech quality and speech-recognition rate compared to conventional methods.},
keywords={},
doi={10.1587/transfun.E99.A.777},
ISSN={1745-1337},
month={April},}

Copy

TY - JOUR
TI - Impact and High-Pitch Noise Suppression Based on Spectral Entropy
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 777
EP - 787
AU - Arata KAWAMURA
AU - Noboru HAYASAKA
AU - Naoto SASAOKA
PY - 2016
DO - 10.1587/transfun.E99.A.777
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E99-A
IS - 4
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - April 2016
AB - We propose an impact and high-pitch noise-suppression method based on spectral entropy. Spectral entropy takes a large value for flat spectral amplitude and a small value for spectra with several lines. We model the impact noise as a flat spectral signal and its damped oscillation as a high-pitch periodic signal consisting of spectra with several lines. We discriminate between the current noise situations by using spectral entropy and adaptively change the noise-suppression parameters used in a zero phase-based impact-noise-suppression method. Simulation results show that the proposed method can improve the perceptual evaluation of the speech quality and speech-recognition rate compared to conventional methods.
ER -