Local and Global Feature Selection for Prosodic Classification of the Word’s Uses

Abdenour Hacine-Gharbi, Philippe Ravier, François Nemo

Abstract

The aim of this study is to evaluate the ability of local or global prosodic features in achieving a classification task of word’s uses. The use of French word “oui” in spontaneous discourse can be identified as belonging to the class “convinced (CV)”or “lack of conviction (NCV)”. Statistics of classical prosodic patterns are considered for the classification task. Local features are those computed on single phonemes. Global features are computed on the whole word. The results show that 10 features completely explain the two clusters CV and NCV carried out by linguistic experts, the features having being selected thanks to the Max-Relevance Min-Redundancy filter selection strategy. The duration of the phoneme /w/ is found to be highly relevant for all the investigated classification systems. Local features are predominantly more relevant than global ones. The system was validated by building classification systems in a speaker dependent mode and in a speaker independent mode and also by investigating manual phoneme segmentation and automatic phoneme segmentation. In the most favorable case (speaker dependent mode and manual phoneme segmentation), the rate reached 87.72%. The classification rate reached 78.57% in the speaker independent mode with automatic phoneme segmentation which is a system configuration close to an industrial one.

References

  1. Boersma, P. and Weenink, D., 2014. Praat: doing phonetics by computer. [Online] (5.3.75) Available at: www.praat.org [Accessed 2014].
  2. Brown, G., Pocock, A., Zhao, M.-J. and Lujan, M., 2012. Conditional Likelihood Maximisation: A Unifying Framework for information theoretic feature selection. Journal of Machine Learning Research, 13(1), pp.27- 66.
  3. Cover, T. and Thomas, J., 1991. Elements of information theory. New York: Wiley Series in telecommunications.
  4. Hacine-Gharbi, A. et al., 2013. A new histogram-based estimation technique of entropy and mutual information using mean squared error minimization. Computers and Electrical Engineering, 39(3), pp.918-33.
  5. Hacine-Gharbi, A., Petit, M., Ravier, P. and Nemo, F., 2015. Prosody Based Automatic Classification of the Uses of French 'oui' as Convinced or Unconvinced Uses. In 4th International Conference on Pattern Recognition Applications and Methods (ICPRAM). Lisboa, Portugal, 2015.
  6. Jain, A., Duin, R. and Mao, J., 2000. Statistical pattern recognition: a review. Trans. Pattern Analysis and Machine Intelligence, 22, (1), pp.4-37.
  7. Kompe, R., 1997. Prosody in Speech Understanding Systems. LNAI.
  8. Manganaro, L., Peskin, B. and Shriberg, E., 2002. Using prosodic and lexical information for speaker. In ICASSP., 2002.
  9. Mary, L. and Yegnanarayana, B., 2008. Extraction and representation of prosodic features for language and and speaker recognition. Speech Communication, pp.782-96.
  10. Petit, M., 2009. Discrimination prosodique et représentation du lexique: application aux emplois des connecteurs discursifs. thesis. PhD Thesis, University of Orléans.
  11. Wang, C., 2001. Prosodic modelling for improved speech recognition and understanding. PhD Thesis, Massachussetts Institute of Technology.
  12. Young, S., Kershaw, D., Odell, J. and Ollason, D., 1999. The HTK Book. Cambridge: Entropic Ltd.
Download


Paper Citation


in Harvard Style

Hacine-Gharbi A., Ravier P. and Nemo F. (2017). Local and Global Feature Selection for Prosodic Classification of the Word’s Uses . In Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-222-6, pages 711-717. DOI: 10.5220/0006251407110717


in Bibtex Style

@conference{icpram17,
author={Abdenour Hacine-Gharbi and Philippe Ravier and François Nemo},
title={Local and Global Feature Selection for Prosodic Classification of the Word’s Uses},
booktitle={Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2017},
pages={711-717},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006251407110717},
isbn={978-989-758-222-6},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Local and Global Feature Selection for Prosodic Classification of the Word’s Uses
SN - 978-989-758-222-6
AU - Hacine-Gharbi A.
AU - Ravier P.
AU - Nemo F.
PY - 2017
SP - 711
EP - 717
DO - 10.5220/0006251407110717