COEVOLUTIONARY ASYMPTOTIC GENETIC ALGORITHM FOR SYLLABLE MODELBASED IDENTIFICATION OF SENTENCES IN AUTOMATIC RECOGNITIONOF CONTINUOUS SPEECH


Cite item

Full Text

Abstract

The last stage of the syllable based automatic continuous speech recognition is the concatenation of the recognized
syllables into a sequence of words. A method for solving this problem is suggested and investigated in the article. It is
based on a specially designed stochastic optimization algorithm which is able to determine the most likely sentence corresponding
to acoustic signal within an acceptable time frame.

References

  1. Unlimited vocabulary speech recognition based on morphs discovered in an unsupervised manner / V. Siivola, T. Hirsimäki, M. Creutz, M. Kurimo // Proc. of the 8th European Conf. on Speech Communication and Technology (Eurospeech). Geneva, Switzerland, 2003. P. 2293-2296.
  2. Rabiner L., Juang B.-H. Fundamentals of Speech Recognition. М. : Prentice Hall, 1993.
  3. Карпов А. А. Модели и программная реализация распознавания русской речи на основе морфемного анализа : дис. … канд. техн. наук. СПб., 2007.
  4. Зализняк А. А. Грамматический словарь русского языка: словоизменение. 4-е изд., испр. и доп. М. : Рус. слов., 2003.
  5. Кипяткова И. С., Карпов А. А. Модуль фонематического транскрибирования для системы распознавания разговорной русской речи // Искусств. интеллект. 2008. № 4. P. 747-757.
  6. Холоденко А. Б. О построении статистических языковых моделей для систем распознавания русской речи // Интеллектуал. системы. 2001. Т. 6. Вып. 1-4. С. 381-394.
  7. Wagner R. A., Fischer M. J. The string-to-string correction problem // J. ACM. 1974. Vol. 21. № 1. P. 168-173.
  8. Knuth D. E. The art of computer programming. Vol. 3. Sorting and Searching. Reading, Mass. : Addison- Wesley, 1973.
  9. Kukich K. Techniques for automatically correcting words in text // ACM Computing Surveys. 1992. Vol. 24. № 4. P. 377-439.
  10. Eiben A., Hinterding R., Michalewicz Z. Parameter control in evolutionary algorithms // IEEE Trans. Evolutionary Algorithms. 1999. Vol. 3. P. 124-141.
  11. Antamoshkin A., Semenkin E. Local Search Efficiency when Optimizing Unimodal Pseudoboolean Functions // Informatica. 1998. Vol. 9. № 3. P. 279-296.
  12. Бежитский С. С., Семенкин Е. С., Семенкина О. Э. Гибридный эволюционный алгоритм для выбора эффективных вариантов систем управления космическими аппаратами // Автоматизация и соврем. технологии. 2005. № 11.
  13. Yakimov Y., Semenkin E., Yakimov I. Hybrid genetic algorithm for a full-profile analysis of XRD powder patterns // Acta Cryst. A64. C226. 2008.
  14. Schlierkamp-Voosen D., Mühlenbein H. Strategy adaptation by competing subpopulations // Parallel Problem Solving from Nature III. Springer-Verlag, 1994.
  15. Potter M. A., De Jong K. A. Cooperative coevolution: an architecture for evolving coadapted subcomponents // Trans. Evolutionary Computation. 2000. Vol. 8. P. 1-29.
  16. Rosin C., Belew R. New Methods for Competitive Coevolution // Trans. Evolutionary Computation. 1997.
  17. Sergienko R., Semenkin E. Competitive Cooperation for Strategy Adaptation in Coevolutionary Genetic Algorithm for Constrained Optimization // IEEE World Congress on Computational Intelligence (WCCI'2010). Barcelona, Spain, 2010. P. 1626-1631.
  18. Семенкин Е. С., Сопов Е. А. Вероятностные эволюционные алгоритмы для оптимизации сложных систем // Сб. тр. междунар. конф. «Intelligent systems» (AIS'05) and «Intelligent CAD» (CAD-2005) : в 3 т. Т. 1. М. : Физматлит, 2005.
  19. Galushin P. V., Semenkin E. S. Asymptotic probabilistic genetic algorithm // Vestnik. Sci. J. of Siberian State Aerospace Univ. named after academician M. F. Reshetnev. Vol. 5 (26). 2009. P. 45-49.

Supplementary files

Supplementary Files
Action
1. JATS XML
Download

Copyright (c) 2011 Zablotskiy S.G., Semenkin E.S., Shvets A.V., Zablotskiy S.G., Semenkin E.S., Shvets A.V.

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies