Abstract
In this paper, we apply Evolutionary Algorithms (EA) to evolve Automatic Speech Recognition Systems (ASRSs) in order to adapt them to acoustic environment changes. The general framework relates to the Evolutionary paradigm and it addresses the problem of robustness of speech recognition as a two level process. First, some initial ASRSs based on feedforward Artificial Neural Networks (ANNs) are designed and trained with an initial speech corpus. Second, the ASRSs are tested in Virtual Acoustic Environments (VAEs) in which we playback some speech test data. By using Evolutionary Operators as mutation, crossover and selection, the adaptation of initial ASRSs to a new VAE is achieved. The VAE includes different real world noises and are physical models of real rooms (1 floor, 1 ceiling and 4 walls) thanks to image methods of sound propagation in small rooms.
Preview
Unable to display preview. Download preview PDF.
References
Allen J., B. & Berkley D. A.: Image Method for efficiently simulating small-room acoustics. JASA 65(4):943–950 (1979).
Bateman, D. C., Bye, D. K. & Hunt M. J.: Spectral normalization and other spectral technics for speech recognition in noise. In Proceedings of the IEEE International conference. on Acoustic Speech Signal Processing, (1)241–244. San Francisco (1992).
Belew, R. K., McInerney, J. & Schraudolph, N. N.: Evolving Networks: Using the Genetic Algorithm with Connectionist Learning. In Proc. Second Artificial Life Conference, pages 511–547, New York, (1991). Addison-Wesley
Das, S., Nadas, A., Nahamoo, D. & Picheny, M.: Adaptation techniques for ambient noise and microphone compensation in the IBM Tangora speech recognition system. In Proceedings of the IEEE International Conference On Acoustic Speech Signal Processing. (1)21–23. Adelaide, Australia (1994).
Goldberg, D.E.: Genetic Algorithms in Search, Optimization & Machine Learning. Addison-Wesley Publishing Company, Inc (1989).
Gong, Y.: Speech recognition in noisy environments: A survey, Journal of Speech Communication (1995), 16: 261–291.
Hermansky, H.: Perceptual Linear Predictive (PLP) Analysis of Speech, Journal of Acoustic Society Am (1990), 87(4) 1738–1752.
Holland, H.: Adaptation in Natural and Artificial Systems. The University of Michigan Press (1975).
Junqua, J. C. & Haton, H.: Robustness in Automatic Speech Recognition, Ed Kluwer Academic Publisher (1996).
Kabré, H. & Spalanzani A.: EVERA: A system for the Modeling and Simulation of Complex Systems. In Proceedings of the First International Workshop on Frontiers in Evolutionary Algorithms, FEA'97, 184–188. North Carolina (1997).
Kabré, H.: On the Active Perception of Speech by Robots. IEEE RJ/MFI (Multi-sensor Fusion and Integration for Intelligent Systems), 775–785. Washington D.C (1996).
Wessels, L and Barnard, E. Avoiding False Local Minima by Proper Initialization of Connections. IEEE Transactions on Neural Networks, vol. 3, No 6, (Nov. 1992).
Mansour, D. & Juang, B. H.: A family of distortion measures based upon projection operation for robust speech recognition. IEEE International Acoustic Speech Signal Process, 36–39. New York (1988).
McGurk, H., MacDonald, J.: Hearing Voices and Seeing Eyes, Nature, 264:746–748 (1976).
Mühlenbein, H. & Schlierkamp-Voosen, D.: Evolution as a Computational Process. Lecture Notes in Computer Science, 188–214, Springer, Berlin (1995).
Spears, W.M., De Jong, K.A., Bäck, T., Fogel, D. and De Garis, H.: An Overview of Evolutionary Computation. In Proceedings of the European Conference on Machine Learning (1993), (667) 442–459.
Yuhas, B.P., Goldstein, M.H. & Sejnowski, T.J.: Interpretation of Acoustic and Visual Speech Signal using Neural Networks. IEEE Common Magazine (1989).
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Spalanzani, A., Kabré, H. (1998). Evolution, learning and speech recognition in changing acoustic environments. In: Eiben, A.E., Bäck, T., Schoenauer, M., Schwefel, HP. (eds) Parallel Problem Solving from Nature — PPSN V. PPSN 1998. Lecture Notes in Computer Science, vol 1498. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0056908
Download citation
DOI: https://doi.org/10.1007/BFb0056908
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65078-2
Online ISBN: 978-3-540-49672-4
eBook Packages: Springer Book Archive