Advertisement

Lithuanian Digits Recognition by Using Hybrid Approach by Combining Lithuanian Google Recognizer and Some Foreign Language Recognizers

  • Tomas RasymasEmail author
  • Vytautas Rudžionis
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 538)

Abstract

In this paper we are presenting our results obtained by experimenting with different classification methods which are suitable for creating hybrid speech recognizer. We tried to create Lithuanian digits recognizer which is capable of producing more than 95 % accuracy. Classification methods that were used are: k-Nearest neighbors (5, 11, 15, and 21), Linear Discriminant Analysis, Quadratic Discriminant Analysis, Logistic Regression, Naïve Bayes, Support Vectors classifier. Experiments were taken using five foreign recognizers: English, Russian, two German and Google recognizer for Lithuanian language. Best results were received when Naïve Bayes classifier was used – 97.51 %. Average accuracy of single recognizers: English – 84.9 %, Russian – 80.6 %, German I – 63 %, German II – 81.4 % and Google recognizer – 82.6 %. By using hybrid approach accuracy was increased by 12.61 % compared with best single recognizer result.

Keywords

Hybrid speech recognition Lithuanian digits recognition Foreign language adaptation 

References

  1. 1.
    Maskeliūnas, R., Rudžionis, A., Ratkevičius, K., Rudžionis, V.: Investigation of foreign languages models for lithuanian speech recognition. Electron. Electr. Eng. 3(91), 15–20 (2009)Google Scholar
  2. 2.
    Rudžionis, V., Raškinis, G., Rudžionis, A., Ratkevičius, K.: Comparative analysis of adapted foreign language and native Lithuanian speech recognizers for voice user interface. Electron. Electr. Eng. 19(7), 90–93 (2013)Google Scholar
  3. 3.
    Rudžionis, V., Raškinis, G., Rudžionis, A., Ratkevičius, K., Bartišiūtė, G.: Web services based hybrid recognizer of Lithuanian voice commands. Electron. Electr. Eng. 20(9), 50–53 (2014)Google Scholar
  4. 4.
    Rasymas, T., Rudžionis, V.: Combining multiple foreign language speech recognizers by using neural networks. In: Human Language Technologies – The Baltic Perspective, pp. 33–39. IOS Press, Amsterdam (2014). doi: 10.3233/978-1-61499-442-8-33
  5. 5.
    Kasparaitis, P.: Lithuanian speech recognition using the English recognizer. Informatica 19(4), 505–516 (2008)Google Scholar
  6. 6.
    Rudžionis, V., Ratkevičius, K., Rudžionis, A., Raškinis, G., Maskeliunas, R.: Recognition of voice commands using hybrid approach. In: Skersys, T., Butleris, R., Butkiene, R. (eds.) ICIST 2013. CCIS, vol. 403, pp. 249–260. Springer, Heidelberg (2013)CrossRefGoogle Scholar
  7. 7.
    Huggins-Daines, D., Kumar, M., Chan, A., Block, A.W., Ravishankar, M., Rudnicky, A.I.: Pocketsphinx: a free, real-time continuous speech recognition system for hand-held devices. In: IEEE ICASSP 2006 Proceedings, vol. 1, pp. 185–188 (2006)Google Scholar
  8. 8.
    Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., Duchesnay, E.: Scikit-learn: machine learning in python. J. Mach. Learn. Res. 12, 2825–2830 (2011)MathSciNetzbMATHGoogle Scholar
  9. 9.
    Schultz, T., Waibel, A.: Language-independent and language-adaptive acoustic modeling for speech recognition. Speech Commun. 35(1), 31–52 (2001)CrossRefGoogle Scholar
  10. 10.
    Wang, Z., Schultz, T., Waibel, A.: Comparison of acoustic model adaptation techniques on non-native speech. In: IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), pp. 540–543 (2003)Google Scholar
  11. 11.
    Meneido, H., Neto, J.: Combination of acoustic models in continuous speech recognition hybdrid systems. Proc. Int. Conf. Spoken Lang. Process. 9, 1000–1029 (2000)Google Scholar
  12. 12.
    Bahi, H., Sellami, M.: A hybrid approach for Arabic speech recognition. In: ACS/IEEE International Conference on Computer Systems and Applications (2003)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  1. 1.Kaunas FacultyVilnius UniversityKaunasLithuania

Personalised recommendations