A hybrid descent method for optimal sigmoid filter design

Kit Yan Chan; Sven Nordholm; Siow Yong Low; Pei Chee Yong; Ka Fai Cedric Yiu

doi:10.1109/LSP.2014.2303498

A hybrid descent method for optimal sigmoid filter design

Kit Yan Chan, Sven Nordholm, Siow Yong Low, Pei Chee Yong, Ka Fai Cedric Yiu

Research output: Journal article publication › Journal article › Academic research › peer-review

2 Citations (Scopus)

Abstract

In this letter, a hybrid descent method is used to determine a set of filter parameters for a sigmoid filter which attempts to work under various SNR conditions. It overcomes the limitations of the current sigmoid filters that performs effectively only at a single SNR. Results show that significant improvement in terms of better speech qualities can be achieved by the proposed sigmoid filter when working under various SNR conditions.

Original language	English
Article number	6728670
Pages (from-to)	478-482
Number of pages	5
Journal	IEEE Signal Processing Letters
Volume	21
Issue number	4
DOIs	https://doi.org/10.1109/LSP.2014.2303498
Publication status	Published - 1 Apr 2014

Keywords

Hybrid optimization method
noise suppression
sigmoid function
speech enhancement
various SNR conditions

ASJC Scopus subject areas

Electrical and Electronic Engineering
Signal Processing
Applied Mathematics

More information

10.1109/LSP.2014.2303498

Cite this

@article{7673e9c1339948c59170d3b6feff231d,

title = "A hybrid descent method for optimal sigmoid filter design",

abstract = "In this letter, a hybrid descent method is used to determine a set of filter parameters for a sigmoid filter which attempts to work under various SNR conditions. It overcomes the limitations of the current sigmoid filters that performs effectively only at a single SNR. Results show that significant improvement in terms of better speech qualities can be achieved by the proposed sigmoid filter when working under various SNR conditions.",

keywords = "Hybrid optimization method, noise suppression, sigmoid function, speech enhancement, various SNR conditions",

author = "Chan, {Kit Yan} and Sven Nordholm and Low, {Siow Yong} and Yong, {Pei Chee} and Yiu, {Ka Fai Cedric}",

year = "2014",

month = apr,

day = "1",

doi = "10.1109/LSP.2014.2303498",

language = "English",

volume = "21",

pages = "478--482",

journal = "IEEE Signal Processing Letters",

issn = "1070-9908",

publisher = "IEEE",

number = "4",

}

TY - JOUR

T1 - A hybrid descent method for optimal sigmoid filter design

AU - Chan, Kit Yan

AU - Nordholm, Sven

AU - Low, Siow Yong

AU - Yong, Pei Chee

AU - Yiu, Ka Fai Cedric

PY - 2014/4/1

Y1 - 2014/4/1

N2 - In this letter, a hybrid descent method is used to determine a set of filter parameters for a sigmoid filter which attempts to work under various SNR conditions. It overcomes the limitations of the current sigmoid filters that performs effectively only at a single SNR. Results show that significant improvement in terms of better speech qualities can be achieved by the proposed sigmoid filter when working under various SNR conditions.

AB - In this letter, a hybrid descent method is used to determine a set of filter parameters for a sigmoid filter which attempts to work under various SNR conditions. It overcomes the limitations of the current sigmoid filters that performs effectively only at a single SNR. Results show that significant improvement in terms of better speech qualities can be achieved by the proposed sigmoid filter when working under various SNR conditions.

KW - Hybrid optimization method

KW - noise suppression

KW - sigmoid function

KW - speech enhancement

KW - various SNR conditions

UR - http://www.scopus.com/inward/record.url?scp=84900573507&partnerID=8YFLogxK

U2 - 10.1109/LSP.2014.2303498

DO - 10.1109/LSP.2014.2303498

M3 - Journal article

SN - 1070-9908

VL - 21

SP - 478

EP - 482

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

IS - 4

M1 - 6728670

ER -

A hybrid descent method for optimal sigmoid filter design

Abstract

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this