Abstract
In this paper, we propose a new approach to robust speaker identification using KPCA (kernel principal component analysis). This approach uses ensembles of classifiers (speaker identifiers) to reduce KPCA computation. KPCA enhances the features for each classifier. To reduce the processing time and memory requirements, we select a subset of limited number of samples randomly which is used as estimation set for each KPCA basis. The experimental result shows that the proposed approach shows better accuracy than PCA and GKPCA (greedy KPCA).
Keywords
- classifier ensemble
- greedy kernel PCA
- speaker identification
This is a preview of subscription content, access via your institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification. John Wiley & Sons (2001)
Shawe-Taylor, J., Cristianini, N.: Kernel Methods for Pattern Analysis. Cambridge University Press (2004)
Scholkopf, B., Smola, A., Muller, K.-R.: Kernel Principal Component Analysis. In: Gerstner, W., Hasler, M., Germond, A., Nicoud, J.-D. (eds.) ICANN 1997. LNCS, vol. 1327, pp. 583–588. Springer, Heidelberg (1997)
Kim, M.-S., Yang, I.-H., Yu, H.-J.: Robust Speaker Identification using Greedy Kernel PCA. In: Proceedings of 20th IEEE International Conference on Tools with Artificial Intelligence, pp. 143–146 (2008)
Mashao, D.J., Skosan, M.: Combining classifier decisions for robust speaker identification. Pattern Recognition 39, 147–155 (2006)
Polikar, R.: Ensemble based Systems in Decision Making. Circuits and Systems Magazine (2006)
Reynolds, D.A., Quatieri, T.F., Dunn, R.B.: Speaker Verification Using Adapted Gaussian Mixture Models. Digital Signal Processing 10, 19–41 (2000)
Reynolds, D.A., Rose, R.C.: Robust text-independent speaker identification using Gaussian mixture speaker models. IEEE Transactions on Speech Audio Processing 3(1), 72–83 (1995)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Yang, IH., Kim, MS., So, BM., Kim, MJ., Yu, HJ. (2012). Robust Speaker Identification Using Ensembles of Kernel Principal Component Analysis. In: Corchado, E., Snášel, V., Abraham, A., Woźniak, M., Graña, M., Cho, SB. (eds) Hybrid Artificial Intelligent Systems. HAIS 2012. Lecture Notes in Computer Science(), vol 7208. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-28942-2_7
Download citation
DOI: https://doi.org/10.1007/978-3-642-28942-2_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-28941-5
Online ISBN: 978-3-642-28942-2
eBook Packages: Computer ScienceComputer Science (R0)
