Email this article Speaker Recognition System using Wavelet Transform under Stress Condition
Abstract
in this paper, we introduced a text-depend speaker recognition by using wavelet transform under stressed conditions. Here we compare different feature such as ARC, LAR, LPCC, MFCC, CEP and after comparison we found that LPCC provides best feature. For decompose signal at two levels Discrete Wavelet Transform is used here. Discrete Wavelet Transform (DWT) based Linear Predictive Cepstral Coefficients (LPCC) used as a feature for recognized the speaker system. For classification Vector Quantization method is used. Four different stressed data has selected for (SUSAS) i.e. stress speech data base for speaker recognition. Improvement is achieved 93% and 94% in case of Lombard and Neutral case.
References
J.P. Campbell, “Speaker Recognition: A Tutorial,”
Proc. IEEE., Vol. 85, no. 9, Sep 1997.
Herman J.M.Steeneken and John.H.L.hasen ‘‘Speech
under stress condition.”IEEE “Robust Speech
process”1999.
Sanjay. Patil“ Speech under stress: Analysis, Modeling
And Recognition,” January 2007.
W.Alkhadi, W.Fakhar and N.Handy ‘‘Automatic
Speech/speaker recognition in noisy environments using
WT,”senior member, IEEE2002.
Shailaja S Yadav, D.G. Bhalke , ‘‘ Speaker
Identification System using Wavelet Transform and
VQ modeling Technique,’’International Journal Of
Computer Applications Volume 112 – No. 9, February
Mahmoud I. Abdalla, Haitham M. Abobakr, Tamer S.
Gaafar ‘‘DWT AndMfccs Based Feature Extraction
Methods For Isolated Word Recognition,’’Volume
– No.20, May 2013 21.
TaabishGulzar, Anand Singh, Sandeep Sharma,
“Comparative Analysis of LPCC, MFCC and BFCC
For the Recognition using Artifical Neural Network,”
International Journal of Computer Application Volume 101-No.12, September 2014.
J.H.L Hasen , Mark A .Clements ‘‘Stresscompensation
noise reduction algorithm for robust speech recognition”
IEEE.
Levent M. Arslan and John H.L Hasen ‘‘Speech
enhancement for cross talk interference” IEEE signal
processing letter, VOL.4, no.4, APRIL 1997.
Murray , I. R, Baber. c and south. A .j “towards
defination and working model of stress and its effects
on Speech.” Speech enhancement communication
vol.20,nov. 1-2,1996.
H.Hermansky, S. Tibrewala, and M. Pavel,“Towards
ASR on Partially Corrupted Speech,” in Proceedings
of ZCSLP, 1996.
J.N.Gowdy, Z.Tufekci ‘‘Mel-scale Discrete Wavelet
Coefficiant for Speech recognition”2000.
Evan Ruzanski1, John H.L. ‘‘ Stress Level
Classification of Speech Using Euclidean Distance
Metricsin a Novel Hybrid Multi-Dimensional Feature
SpaceHansen”2006.
Alm,C.O., Roth, D., Sproat, R.: Emotions from Text:
‘‘Machine Learning for Text based Emotion Prediction. In: Proceedings of HLT/EMNLP 05, Vancouver’’ (2005).
A. Mantilla-Caeiros, M. Nakano-Miyatake, H. Perez-
Meana, “A New Wavelet Function for Audio and
Speech Processing”, 50th MWSCAS, pp. 101-
(2007).
HynekBoˇril, Member, IEEE, and John H. L. Hansen,
Fellow, IEEE ‘‘Unsupervised Equalization of
Lombard Effect for Speech Recognition in Noisy
Adverse Environments” Transactions On Audio,
Speech, And Language Processing, Vol. 18, No. 6,
August 2010.
RuhiSarikaya and John N. Gowdy‘‘ Wavelet Based
Analysis of Speech Under Stress” Digital Speech and
Refbacks
- There are currently no refbacks.
Copyright © 2013, All rights reserved.| ijseat.com
International Journal of Science Engineering and Advance Technology is licensed under a Creative Commons Attribution 3.0 Unported License.Based on a work at IJSEat , Permissions beyond the scope of this license may be available at http://creativecommons.org/licenses/by/3.0/deed.en_GB.