Abstract
Some well known theoretical results concerning the universal approximation property of MLP neural networks with one hidden layer have shown that for any function f from [0,1]n to \(\mathcal{R}\), only the output layer weights depend on f. We use this result to propose a network architecture called the predictive Kohonen map allowing to design a new speech features extractor. We give experimental results of this approach on a phonemes recognition task.
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
An erratum to this chapter can be found at http://dx.doi.org/10.1007/11550907_163 .
This is a preview of subscription content, log in via an 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
Hermansky, H.: Should recogizers have ears? Speech Communication 25, 3–27 (1998)
Lapedes, A., Farber, R.: Nonlinear signal processing using neural networks: Prediction and system modelling. Internal Report, Los Alamos National Laboratory (1987)
Thyssen, J., Nielsen, H., Hansen, S.D.: Non-linear short-term prediction in speech coding. In: Proc. of Int. Conf. on Signal and Speech Processing, vol. 1, pp. 185–188 (1994)
Reichl, W., Harengel, S., Wolferstetter, F., Ruske, G.: Neural networks for nonlinear discriminant analysis in continuous speech recognition. In: Eurospeech, pp. 537–540 (1995)
Hunt, M.J., Lefebvre, C.: A comparison of several acoustic representations for speech recognition with degraded and undegraded speech. In: Int. Conf. on Speech and Signal Processing, vol. 2, pp. 262–265 (1989)
Biem, A., Katagiri, S.: Filter bank design based on Discriminative Feature Extraction. In: Proc. of Int. Conf. on Signal and Speech Processing, vol. 1, pp. 485–488 (1994)
Gas, B., Zarader, J.L., Chavy, C., Chetouani, M.: Discriminant neural predictive coding applied to phoneme recognition. Neurocomputing 56, 141–166 (2004)
Chetouani, M., Faundez-Zanuy, M., Gas, B., Zarader, J.L.: A New Nonlinear speaker parameterization algorithm for speaker identification. In: Proc. of ISCA Tutorial and Research Workshop on Speaker and Recognition Langage Workshop, pp. 309–314 (2004)
Hecht-Nielsen, R.: Kolmogorov’s mapping neural network existence theorem. In: Proc. of Int. Conf. on Neural Networks, pp. 11–13 (1987)
Girosi, F., Poggio, T.: Representation properties of networks: Kolmogorov’s theorem is irrelevant. Neural Computation 1(4), 465–469 (1989)
Hornik, K.: Multilayer feedforward networks are universal approximators. Neural Networks, 2, 359–366 (1089)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Gas, B., Chetouani, M., Zarader, JL., Charbuillet, C. (2005). Predictive Kohonen Map for Speech Features Extraction. In: Duch, W., Kacprzyk, J., Oja, E., Zadrożny, S. (eds) Artificial Neural Networks: Formal Models and Their Applications – ICANN 2005. ICANN 2005. Lecture Notes in Computer Science, vol 3697. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11550907_125
Download citation
DOI: https://doi.org/10.1007/11550907_125
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-28755-1
Online ISBN: 978-3-540-28756-8
eBook Packages: Computer ScienceComputer Science (R0)