KLI

적응필터와 음성구간 검출을 이용한 소음감소

Metadata Downloads
Alternative Title
Noise Reduction using Adaptive Filter and Voice Interval Detection
Abstract
본 논문은 소음과 음성이 섞인 신호에서 소음을 효과적으로 제거하여 음질개선을 도모하는 시스템을 제안한다. 제안된 시스템은 실제 산업현장에서의 음성과 소음이 섞인 신호를 LMS 알고리듬을 이용한 적응필터와 음성신호에 적합한 저주파 대역필터를 통과시킨다. 이 신호를 음성구간과 비 음성구간을 검출하는 알고리듬에 적용하여 비 음성구간의 소음을 제거함으로서 음질을 개선하였다.
We propose the new system which consists of adaptive filter using least mean square algorithm, lowpass filter(LPF) and part of voice interval detection in order to improve voice quality by reducing the noise components in the industry fields. The mixed signal with voice and noise is passed through the adaptive filter to remove the background noise of the original input signal. We removed the high frequency noise components by passing through LPF. Finally, we could obtain the good quality of voice from the proposed system close to original voice by removing the noise between the voice intervals.
We propose the new system which consists of adaptive filter using least mean square algorithm, lowpass filter(LPF) and part of voice interval detection in order to improve voice quality by reducing the noise components in the industry fields. The mixed signal with voice and noise is passed through the adaptive filter to remove the background noise of the original input signal. We removed the high frequency noise components by passing through LPF. Finally, we could obtain the good quality of voice from the proposed system close to original voice by removing the noise between the voice intervals.
Author(s)
정의필
Issued Date
2001
Type
Research Laboratory
URI
https://oak.ulsan.ac.kr/handle/2021.oak/3939
http://ulsan.dcollection.net/jsp/common/DcLoOrgPer.jsp?sItemId=000002024698
Alternative Author(s)
Chong, Uipil
Publisher
공학연구논문집
Language
kor
Rights
울산대학교 저작물은 저작권에 의해 보호받습니다.
Citation Volume
32
Citation Number
1
Citation Start Page
275
Citation End Page
283
Appears in Collections:
Research Laboratory > Engineering Research
Authorize & License
  • Authorize공개
Files in This Item:

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.