Research Article
PCA, SFS or LDA: What is the best choice for extracting speaker features
|
International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
|
| Volume 15 - Issue 3 |
| Published: February 2011 |
| Authors: Abdelghani Harrag, Tayeb Mohamadi |
10.5120/1932-2578
|
Abdelghani Harrag, Tayeb Mohamadi . PCA, SFS or LDA: What is the best choice for extracting speaker features. International Journal of Computer Applications. 15, 3 (February 2011), 1-3. DOI=10.5120/1932-2578
@article{ 10.5120/1932-2578,
author = { Abdelghani Harrag,Tayeb Mohamadi },
title = { PCA, SFS or LDA: What is the best choice for extracting speaker features },
journal = { International Journal of Computer Applications },
year = { 2011 },
volume = { 15 },
number = { 3 },
pages = { 1-3 },
doi = { 10.5120/1932-2578 },
publisher = { Foundation of Computer Science (FCS), NY, USA }
}
%0 Journal Article
%D 2011
%A Abdelghani Harrag
%A Tayeb Mohamadi
%T PCA, SFS or LDA: What is the best choice for extracting speaker features%T
%J International Journal of Computer Applications
%V 15
%N 3
%P 1-3
%R 10.5120/1932-2578
%I Foundation of Computer Science (FCS), NY, USA
Abstract
Feature extraction is the process of deriving new weakly correlated features from the original features in order to reduce the cost of feature measurement, increase classifier efficiency, and allows higher classification accuracy. The selection and quality of the features representing each pattern have considerable bearing on the success of subsequent pattern classification. In this paper, we supply a comparative study for best feature extraction method for speaker recognition system. A Linear Discriminant Analysis (LDA) method is compared to two well-known feature extraction techniques, namely Principal Component Analysis (PCA) and Sequential Forward Search (SFS). Evaluation is carried out on Arabic speech database using four acoustic representations combined with prosodic features. We show that LDA-based feature outperformed PCA and SFS in acoustic alone as well as for acoustic and prosodic combined features.
References
- Przybocki, M. A. and Martin, A. F. 2004. NIST Speaker Recognition Evaluation Chronicles, Odyssey, Toledo Espana.
- Harrag A., Mohamadi T. , Serignat J.F. 2005. LDA Combination of Pitch and MFCC Features in Speaker Recognition, INDICON Chennai, India, (Dec 2005). pp. 237-240.
- K. Sri Rama Murty and B. Yegnanarayana, Combining evidence from residual phase and MFCC features for speaker recognition, IEEE Signal Processing Letters, 13 (2006), pp. 52-55.
- Yegnanarayana, B., Prasanna, S.R.M., Zachariah, J.M., and Gupta, C.S. 2005. Combining evidence from source, suprasegmental and spectral features for a fixed-text speaker verification system, IEEE Trans Speech and Audio Processing, (2005), 13, pp. 575-582.
- Avinash B., Guruprasad S. and Yegnanarayana B. 2010. Exploring subsegmental and suprasegmental features for a text-dependent speaker verification in distant speech signals, Interspeech 2010, Makuhari, Chiba, Japan. pp. 1073-1076.
- Chakroborty, S. and Saha, G. 2009. Improved Closed set Text-Independent Speaker Identification by Combining MFCC with Evidence from Flipped Filter Banks. International Journal of Signal Processing, (2009), 5, 1-2, pp. 11-19.
- Rabiner, L. and Huang, B. H. 1993. Fundamentals of Speech Recognition, Prentice Hall, Englewood Cliffs.
- Harrag A. and Mohamadi T. 2010. QSDAS: New Quranic Speech Database for Arabic Speaker Recognition, AJSE Journal, Theme Issue on Arabic Computing, Vol. 35(2C), (December 2010), pp. 7-19.
- Whitney, A. 1997. A direct method of nonparametric measurement selection, IEEE Trans. Comput., (1997), 20, pp. 1100-1103.
- Jolliffe, I.T. 2002. Principal Component Analysis, Springer.
- Duda, R.O., Hart, P.E., Stork, D.G. 2000. Pattern Classification, Wiley Interscience.
Index Terms
Keywords