Automatic Classification of Arabic Back Consonants for Correction of Phonemic Substitution
Article Sidebar
Published:
Jun 28, 2014
Main Article Content
Abstract
In this article, we present an automatic classification system for Arabic back consonants [q][ɂ] and [x][ḥ]. The main goal is to build a speech therapy aid system for Algerian children suffering from functional phonemic substitution problems. This system is an application of Automatic Speech Recognition (ASR) based on the Hidden Markov Model or HMM (Hidden Markov Model). The parameterization of speech signals is based on a cepstral representation using MFCC (Mel Frequency Cepstral Coefficients) with its first and second derivatives. An objective assessment was applied to our work. The latter shows good performance, with Recognition Rates of 90.71% for the sound [q] and 91.71% for the [x].
Article Details
How to Cite
Abed, A., & GUERTI, M. (2014). Automatic Classification of Arabic Back Consonants for Correction of Phonemic Substitution. AL-Lisaniyyat, 20(1), 127-134. https://doi.org/10.61850/allj.v20i1.511
Section
Articles
References
7. Références
[1] Abed, A. and Guerti, M., “Application des HMM à la substitution Phonémique dans l'Arabe Parlé”, Journées d'Etudes Algéro-Françaises de Doctorants en Si- gnal-Image & Applications, JEAFD'2012, Alger, 18-23, 2012. [2] Abed, A. and Guerti, M., "Errors Classi- fication of Phonemic Substitution in Ara134
bic Speech”, International Congress on
Telecommunication and Application’14. Bejaia, Algeria, 23-24 Apr 2014. [3] Amrouche, A., Debyeche, M., Ahmed, AT. Rouvaen, TM. and Yagoub, M.C.E,, "An efficient speech recognition system in adverse conditions using the non parametric regression”, Engineering Applications of Artificial Intelligence, Elsevier, 23: 85-94, 2010. [4] Kumari, R.S.S., Nidhyananthan, S.S. and
Anand, G., ”Fused mel feature sets based text-independent speaker identification using gaussian mixture model”, Elsevier Procedia Engineering, editor, International Conference on Communication Technology and System Design, 30: 319- 326, 2012. [5] Zeng, J., Duan, J. and Wu, C., "A new
distance measure for hidden markov models”, Expert Systems with Applica- tions, Elsevier, 37:1550-1555, 2010. [6] Bilmes, J.A., "A gentle Tutorial of the EM Algorithm and its applications to Parameter Estimation for Gaussian Mixture
and Hidden Markov Models”, Technical report, ICSI-TR-97-021, 1998. [7] Davis, S. P. and Mermelstein, P., "Com- parison of parametric representations for monosyllabic word recognition in contin- uously spoken sentences”, IEEE Transac- tions on Acoustics, Speech, and Signal Processing, 28: 357-366, 1980. [8] Hacine-Gharbi, A., “Sélection de para- mètres acoustiques pertinents pourla reconnaissance de la parole”, Thèse de doctorat, université de Sétif Algérie, 2012. [9] Haton, I.P., Cerisara, C., Fohr, D, Laprie,
Y. and Smaili, K., ”Reconnaissance au- tomatique de la parole : du signal à son interprétation”, Paris: Dunod, 2006. [10] Rabiner, LR. and Juan, B.H., “Funda- mentals of speech recognition”, Eng- lewood Cliffs, N.J., USA: Prentice Hall, 1993.
[1] Abed, A. and Guerti, M., “Application des HMM à la substitution Phonémique dans l'Arabe Parlé”, Journées d'Etudes Algéro-Françaises de Doctorants en Si- gnal-Image & Applications, JEAFD'2012, Alger, 18-23, 2012. [2] Abed, A. and Guerti, M., "Errors Classi- fication of Phonemic Substitution in Ara134
bic Speech”, International Congress on
Telecommunication and Application’14. Bejaia, Algeria, 23-24 Apr 2014. [3] Amrouche, A., Debyeche, M., Ahmed, AT. Rouvaen, TM. and Yagoub, M.C.E,, "An efficient speech recognition system in adverse conditions using the non parametric regression”, Engineering Applications of Artificial Intelligence, Elsevier, 23: 85-94, 2010. [4] Kumari, R.S.S., Nidhyananthan, S.S. and
Anand, G., ”Fused mel feature sets based text-independent speaker identification using gaussian mixture model”, Elsevier Procedia Engineering, editor, International Conference on Communication Technology and System Design, 30: 319- 326, 2012. [5] Zeng, J., Duan, J. and Wu, C., "A new
distance measure for hidden markov models”, Expert Systems with Applica- tions, Elsevier, 37:1550-1555, 2010. [6] Bilmes, J.A., "A gentle Tutorial of the EM Algorithm and its applications to Parameter Estimation for Gaussian Mixture
and Hidden Markov Models”, Technical report, ICSI-TR-97-021, 1998. [7] Davis, S. P. and Mermelstein, P., "Com- parison of parametric representations for monosyllabic word recognition in contin- uously spoken sentences”, IEEE Transac- tions on Acoustics, Speech, and Signal Processing, 28: 357-366, 1980. [8] Hacine-Gharbi, A., “Sélection de para- mètres acoustiques pertinents pourla reconnaissance de la parole”, Thèse de doctorat, université de Sétif Algérie, 2012. [9] Haton, I.P., Cerisara, C., Fohr, D, Laprie,
Y. and Smaili, K., ”Reconnaissance au- tomatique de la parole : du signal à son interprétation”, Paris: Dunod, 2006. [10] Rabiner, LR. and Juan, B.H., “Funda- mentals of speech recognition”, Eng- lewood Cliffs, N.J., USA: Prentice Hall, 1993.