SVM-Based Detection of Misannotated Words in Read Speech Corpora

  • Jindřich Matoušek
  • Daniel Tihelka
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8082)

Abstract

Automatic detection of misannotated words in single-speaker read-speech corpora is investigated in this paper. Support vector machine (SVM) classifier was proposed to detect the misannotated words. Its performance was evaluated with respect to various word-level feature sets. The SVM classifier was shown to perform very well with both high precision and recall scores and with F1 measure being almost 88%. This is a statistically significant improvement over a traditionally used outlier-based detection method.

Keywords

annotation error detection classification support vector machine read speech corpora 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Matoušek, J., Tihelka, D., Šmídl, L.: On the impact of annotation errors on unit-selection speech synthesis. In: Sojka, P., Horák, A., Kopeček, I., Pala, K. (eds.) TSD 2012. LNCS, vol. 7499, pp. 456–463. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  2. 2.
    Matoušek, J., Romportl, J.: Recording and Annotation of Speech Corpus for Czech Unit Selection Speech Synthesis. In: Matoušek, V., Mautner, P. (eds.) TSD 2007. LNCS (LNAI), vol. 4629, pp. 326–333. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  3. 3.
    Adell, J., Agüero, P.D., Bonafonte, A.: Database pruning for unsupervised building of text-to-speech voices. In: Proc. ICASSP, Toulouse, France, pp. 889–892 (2006)Google Scholar
  4. 4.
    Tachibana, R., Nagano, T., Kurata, G., Nishimura, M., Babaguchi, N.: Preliminary experiments toward automatic generation of new TTS voices from recorded speech alone. In: Proc. INTERSPEECH, Antwerp, Belgium, pp. 1917–1920 (2007)Google Scholar
  5. 5.
    Wei, S., Hu, G., Hu, Y., Wang, R.H.: A new method for mispronunciation detection using support vector machine based on pronunciation space models. Speech Commun. 51(10), 896–905 (2009)CrossRefGoogle Scholar
  6. 6.
    Kominek, J., Black, A.: Impact of durational outlier removal from unit selection catalogs. In: Proc. SSW, Pittsburgh, USA, pp. 155–160 (2004)Google Scholar
  7. 7.
    Lu, H., Wei, S., Dai, L., Wang, R.H.: Automatic error detection for unit selection speech synthesis using log likelihood ratio based SVM classifier. In: Proc. INTERSPEECH, Makuhari, Japan, pp. 162–165 (2010)Google Scholar
  8. 8.
    Wang, W.Y., Georgila, K.: Automatic detection of unnatural word-level segments in unit-selection speech synthesis. In: Proc. ASRU, Hawaii, USA, pp. 289–294 (2011)Google Scholar
  9. 9.
    Tihelka, D., Kala, J., Matoušek, J.: Enhancements of Viterbi search for fast unit selection synthesis. In: Proc. INTERSPEECH, Makuhari, Japan, pp. 174–177 (2010)Google Scholar
  10. 10.
    Young, S., Evermann, G., Gales, M., Hain, T., Kershaw, D., Liu, X., Moore, G., Odell, J., Ollason, D., Povey, D., Valtchev, V., Woodland, P.: HTK Book (for HTK Version 3.4). The Cambridge University, Cambridge (2006)Google Scholar
  11. 11.
    Matoušek, J., Tihelka, D., Psutka, J.V.: Experiments with Automatic Segmentation for Czech Speech Synthesis. In: Matoušek, V., Mautner, P. (eds.) TSD 2003. LNCS (LNAI), vol. 2807, pp. 287–294. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  12. 12.
    Matoušek, J., Romportl, J.: Automatic pitch-synchronous phonetic segmentation. In: Proc. INTERSPEECH, Brisbane, Australia, pp. 1626–1629 (2008)Google Scholar
  13. 13.
    Dietterich, T.G.: Approximate statistical tests for comparing supervised classification learning algorithms. Neural Comput. 10, 1895–1923 (1998)CrossRefGoogle Scholar
  14. 14.
    Cortes, C., Vapnik, V.: Support-vector networks. Machine Leaming 20(3), 273–279 (1995)MATHGoogle Scholar
  15. 15.
    Matoušek, J., Tihelka, D.: Annotation errors detection in TTS corpora. In: Proc. Interspeech, Lyon, France (2013)Google Scholar
  16. 16.
    Romportl, J., Kala, J.: Prosody modelling in Czech text-to-speech synthesis. In: Proc. SSW, Bonn, Germany, pp. 200–205 (2007)Google Scholar
  17. 17.
    Taylor, P., Caley, R., Black, A., King, S.: Edinburgh speech tools library: System documentation (1999), http://www.cstr.ed.ac.uk/projects/speech_tools/manual-1.2.0/
  18. 18.
    Pedregosa, F., Varoquaux, G., Gramfort, A., Thirion, V.M.B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perror, M.: Édouard Duchesnay: Scikit-learn: Machine learning in Python. J. Machine Learn. Res. 12, 2825–2830 (2011)Google Scholar
  19. 19.
    Přibil, J., Přibilová, A.: Comparison of spectral and prosodic parameters of male and female emotional speech in Czech and Slovak. In: Proc. ICASSP, Prague, Czech Republic, pp. 4720–4723 (2011)Google Scholar
  20. 20.
    Ircing, P., Psutka, J., Psutka, J.V.: Using morphological information for robust language modeling in Czech ASR system. IEEE Trans. Audio Speech Lang. Process. 17, 840–847 (2009)CrossRefGoogle Scholar
  21. 21.
    Psutka, J., Švec, J., Psutka, J.V., Vaněk, J., Pražák, A., Šmídl, L., Ircing, P.: System for fast lexical and phonetic spoken term detection in a Czech cultural heritage archive. EURASIP J. Audio Speech Music Process. 10 (2011)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Jindřich Matoušek
    • 1
  • Daniel Tihelka
    • 1
  1. 1.Faculty of Applied Sciences, Dept. of CyberneticsUniversity of West BohemiaPlzeňCzech Republic

Personalised recommendations