The objective of Speech Enhancement is to improve
the quality of audio signal that are degraded speech signal used
in various application i.e. VOIP, hearing aids, teleconferencing
systems. Currently the Speech Enhancement deals with several
problems to improve requirements of speech recognition such as
background noise that is presented in speech signal. The
proposed work introduces alternative ways for providing best
quality of speech signal aimed at speech enhances using
Modified Spectral Subtraction algorithms (MSS). The scope of
the proposed solution is to improve the SNR value of audio
signal as well as to reduce completely background noise without
affecting signal power. This paper defines the modified spectral
subtraction method to estimate noise in non-stationary speech
environment. Finally, we have discussed some of the
performance parameters such as: time required as number of
samples increases and Improved SNR.
Published In : IJCAT Journal Volume 1, Issue 4
Date of Publication : 31 May 2014
Pages : 121 - 124
Figures : 05
Tables : --
Publication Link : Speech Enhancement using Modified Spectral
Subtraction Algorithm
Vijay Kumar M A : Asst. Prof, Department of ECE, VTU, PESIT&M
Shivamogga, 577204, India
Kishore T K : Student Department of ECE, VTU, PESIT&M
Shivamogga, 577204, India
Karthik S : Student Department of ECE, VTU, PESIT&M
Shivamogga, 577204, India
Gagan H R : Student Department of ECE, VTU, PESIT&M
Shivamogga, 577204, India
Speech Enhancement
Spectral Subtraction
SNR
This proposed work presents a method to improve speech
enhancing based on modified spectral subtraction
algorithm was introduced. For effective noise reduction
with minimal distortion proposed algorithm takes in
account perceptual aspects of human ear. It can be seen
from the experimental results that proposed method
effectively reduces background noise. The scope of the
proposed solution is to improve the SNR value of audio
signal as well as to reduce completely background noise
without affecting signal power.
[1] Ekaterina Verteletskaya, Boris Simak, ‘Noise Reduction
Based on Modified Spectral Subtraction Method in
IAENG International Journal of Computer Science, 38:1,
IJCS_38_1_10, published in 2011
[2] Zhixin Chen ILX Lightwave Corporation, Bozeman,
Montana, USA. Simulation of Spectral Subtraction
Based Noise Reduction Method, (IJACSA) in 2011
[3] Anuradha R. Fukane, Shashikant L. Sahare, Noise
estimation Algorithms for Speech Enhancement in highly
non-stationary Environments , published in 2011
[4] Bo Li1, Yu Tsao, Khe Chai Sim, An Investigation of
Spectral Restoration Algorithms for Deep Neural
Networks based Noise Robust Speech Recognition in
2009
[5] Yukihiro NOMURAy, Jianming LUy, Hiroo SEKIYAy,
and Takashi YAHAGI , Spectral Subtraction Based On
Speech/Noise-Dominant Classification at International
Workshop on Acoustic Echo and Noise Control, Kyoto,
Japan, Sept. 2003.
