Learning Phonetic Rules in a Speech
Recognition System

Zoltn Alexin1, Jnos Csirik2, Tibor Gyimthy3,
Mark Jelasity3 and Lszl Tth3

1 Department of Applied Informatics, J6zsef Attila University

rpd tr 2, H-6720 Szeged, Hungary
Phone: (36) +(62) 454293, Fax: (36) +(62) 312292
e-mail: alexin@inf.u-szeged.hu

2 Department of Computer Science, Jzsef Attila University

rpd tr 2, H-6720 Szeged, Hungary
Phone: (36) +(62) 454370, Fax: (36) +(62) 312292
e-mail: csirik@inf.u-szeged.hu

3 Research Group on Artificial Intelligence
Hungarian Academy of Sciences,
Aradi vrtanuk tere 1, H-6720 Szeged, Hungary
Phone: (36) +(62) 454139, Fax: (36) +(62) 312508
e-mail: gyimi, jelasity, tothl@inf.u-szeged.hu


Abstract. Current speech recognition systems can be categorized into
two broad classes; the knowledge-based approach and the stochastic one.
In this paper we present a rule-based method for the recognition of Hungarian 
vowels. A spectrogram model was used as a front-end module
and some acoustic features were extracted (e.g. locations, intensities and
shapes of local maxima) from spectrograms by using a genetic algorithm
method. On the basis of these features we developed a rule set for the
recognition of isolated Hungarian vowels. These rules represented by Prolog 
clauses were refined by the IMPUT Inductive Logic Programming
method.
References

1.	Alexin, Z., Gyim6thy, T., Bostrm, H.: Integrating Algorithmic Debugging and
Unfolding Transformation in an Interactive Learner in Proceedings of the 12th
European Conference on Artificial Intelligence ECAI-96 ed. Wolfgang Wahlster,
Budapest, Hungary (1996) 403407 John Wiley & Sons Ltd. (1996)
2.	Bolla, Klmn: Magyar fonetikai atlasz, A szegmentls hangszerkezet elemei
Nemzeti Tanknyvkiad Rt Budapest (1995) in Hungarian
3.	Cooke, M. (eds.): Visual Representations of Speech Signals, Wiley, 1993
4.	Fohr, D., Haton, J., Laprie, Y.: Knowledge-Based Techniques in Acoustic-Phonetic
Decoding of Speech: Interest and Limitations, International Journal on Pattern
Recognition and Artificial Intelligence Vol. 8, No. 1, (1994) 133153
5.	Huebener, K., Carson-Berndsen, J.: Phoneme Recognition Using Acoustic Events,
Verbmobil Technical Report No. 15, June 1994
6.	Jelasity, M., Dombi, J.: GAS, an Approach on Modeling Species in Genetic Algorithms. 
Proceedings of the EA95 (1995)
7.	Kkai, G., Alexin, Z., Kocsis, F.: The IDT System and its Application for Learning
Prolog Programs. Proc. of the Sixth International Conference on Artificial Intelligence 
and Information Control Systems of Robots (AIICSR-94) Smolenice Castle
Slovakia September 12-16. (1994) 315320
8.	Lamel, L. F.: A knowledge-based system for stop consonant identification based on
speech spectrogram reading, Computer Speech and Language (1993) 2, 169191
9.	Lavrac, N., Dzeroski, S.: Inductive Logic Programming: Techniques and Applications 
Ellis Horwood, (1994)
10.	Mammachandran, R. P., Mammone, R. J. (eds.): Modern Methods of Speech Processing, 
Kluwer Academic, 1995
11.	Mercier, G. et al.: Recognition of speaker-dependent continuous speech with KEAL,
IEE Proceedings Vol. 136, Pt.1, No.2, April (1989)
12.	Muggleton, S., De Raedt, L.: Inductive Logic Programming: Theory and methods,
Journal of Logic Programming 19 (20) (1994) 629679
13.	Rabiner, L. R., Schafer, R. W.: Digital Processing of Speech Signals Prentice Hall
(1978)
14.	Waibel, A., Lee, K. (eds.): Readings in Speech Recognition, Morgan Kaufmann
(1990)
15.	Zue, V. W.: The Use of Speech Knowledge in Automatic Speech Recognition, Proc.
IEEE, Vol. 73, No. 11, (1985) 16021615
