Phonetic encoding method in the isolated words recognition problem

Theory and Methods of Signal Processing


A phonetic approach to the problem of automatic recognition of isolated words is investigated. The phonetic encoding method whereby each word from a vocabulary is associated with the code sequence of stable phonemes is proposed. The information-theoretical estimate of vocabulary confusability, the calculations of which rely on the phonetic database of a speaker and the communications channel SNR, is synthesized using the Kullback-Leibler divergence properties. In the experimental study of the proposed method, the mutual influence between the recognition quality and the proposed estimate of confusability is demonstrated by solving the problem of recognition of words in the Russian speech. It is established that the introduced requirement to isolated syllable pronunciation makes it possible to attain the 90–95% accuracy of recognition for vocabularies containing 2000 words.


Recognition Quality Automatic Speech Recognition Speech Rate Subject Field Automatic Speech Recognition System 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    A. V. Kozlov, G. V. Savvina, and V. Yu. Shelepov, Iskusstvennyi Intellekt, No. 1, 156 (2003).Google Scholar
  2. 2.
    V. V. Savchenko, J. Commun. Technol. Electron. 50, 286 (2005).Google Scholar
  3. 3.
    V. Sorokin and A. Tananykin, J. Commun. Technol. Electron. 55, 1542 (2010).CrossRefGoogle Scholar
  4. 4.
    V. V. Savchenko, Izv. Vyssh. Uchebn. Zaved., Radioelektronika, No. 5, 31 (2009).Google Scholar
  5. 5.
    S. Kullback, Information Theory and Statistics (Dover, New York, 1997).MATHGoogle Scholar
  6. 6.
    I. S. Kipyatkova and A. A. Karpov, Tr. SPIIRAN, No. 12, 7 (2010).Google Scholar
  7. 7.
    B. Tan, in Proc. Conf. Electrical Power Systems and Computers, Lecture Notes in Electrical Engineering (Springer-Verlag, New York, 2011), Vol. 99, p. 771.Google Scholar
  8. 8.
    B. Mérialdo, IBM J. Res. Dev. 32, 227 (1988).CrossRefGoogle Scholar
  9. 9.
    J. Anguita, J. Hernando, S. Peillon, and A. Bramoulle, IEEE Signal Process. Lett. 12, 585 (2005).CrossRefGoogle Scholar
  10. 10.
    A. V. Savchenko, Vestn. Komp. Inform. Tekhnol., No. 8, 14 (2012).Google Scholar
  11. 11.
    V. V. Savchenko, J. Commun. Technol. Electron. 42, 393 (1997).Google Scholar
  12. 12.
    A. V. Gerasimov, O. A. Morozov, and V. R. Fidel’man, J. Commun. Technol. Electron. 50, 1192 (2005).Google Scholar
  13. 13.
    S. L. Marple, Jr. Digital Spectral Analysis: with Applications (Prentice-Hall, Englewood Cliffs, N. J., 1987; Mir, Moscow, 1990).Google Scholar
  14. 14.
    Springer Handbook of Speech Recognition, Ed. by J. Benesty, M. Sondh, Y. Huang (Springer-Verlag, New York, 2008).Google Scholar
  15. 15.
    A. I. Tsyplikhin and V. N. Sorokin, Inform. Protsessy 6, 177 (2006).Google Scholar
  16. 16.
    R. Sibson, Comp. J. (British Comp. Soc.) 16(1), 30 (1973).MathSciNetGoogle Scholar
  17. 17.
    A. V. Savchenko, Opt. Memory and Neural Networks (Inform. Opt.) 21, 219 (2012).CrossRefMathSciNetGoogle Scholar
  18. 18.
    V. V. Savchenko and A. V. Savchenko, Sist. Uprav. Inform. Tekhnol., No. 2, 284 (2012).Google Scholar
  19. 19.
    M. Schuster, in Proc. 11th Pacific Rim Int. Conf. on Trends in Artificial Intelligence, Lecture Notes in Comp. Sci. (Springer, New York, 2010), Vol. 6230, p. 8.Google Scholar
  20. 20.
    A. V. Savchenko, Avtom. and Remote Control 74, 1225 (2013).CrossRefGoogle Scholar

Copyright information

© Pleiades Publishing, Inc. 2014

Authors and Affiliations

  1. 1.National Research University Higher School of EconomicsNizhni NovgorodRussia

Personalised recommendations