| Peer-Reviewed

Hybrid Techniques for Arabic Letter Recognition

Received: 30 August 2014     Accepted: 22 January 2015     Published: 2 February 2015
Views:       Downloads:
Abstract

In this paper we investigate the use of the feed-forward back propagation neural networks (FFBPNN) for automatic speech recognition of Arabic letters with their four vowels (Fatha, dhamma, Kasra, Soukoun). This investigation will constitute a basically step for the recognition of continuous Speech. Features were extracted from recorded corpus by using a variety of conventional methods such as Linear Predictive Codes (LPC), Perceptual Linear Prediction (PLP), Relative Spectral Perceptual Linear Prediction (RASTA-PLP), Mel Frequency Cepstral Coefficients (MFCC), Continuous Wavelet Transform (CWT), etc. Here, several hybrid methods have been used too. Since the extracted features have large dimensionalities they were reduced by conserving the most discriminatory information with the Principal Component Analysis (PCA) technique. The recognition performance has been improved particularly when we use the PLP method followed by PCA technique.

Published in International Journal of Intelligent Information Systems (Volume 4, Issue 1)
DOI 10.11648/j.ijiis.20150401.14
Page(s) 27-34
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2015. Published by Science Publishing Group

Keywords

Speech Recognition, Arabic Letters, Hybrid Techniques, MFCC, PLP, LPCC, PCA and FFBPNN

References
[1] Abdulfattah Ahmad M. and El Awady R. M., “Phonetic Recognition of Arabic Alphabet Letters Using Neural Networks,” International Journal of Electric & Computer Sciences IJECS-IJENS, Vol. 11, No. 01, 112501-3434 IJECS-IJENS ©, February 2011.
[2] Al Azzawi Kh. Y. and Daqrouq Kh., “Feedforward Backpropagation Neural Network Method for Arabic Vowel Recognition Based on Wavlet Linear Prediction Coding,” International Journal of Advances in Engeneering & Technologiy”, Ijalet ISSN:2231-1963, Sept. 2011.
[3] Al-Irhaim Y. F. and Saeed E. Gh., “Arabic Word Recognition Using Wavelet Neural Network,” Third Science Conference in Information Technology, November 2010.
[4] Alkhouli M., “Alaswaat Alaghawaiyah,” Daar Alfalah, Jordan, 1990.
[5] Alotaibi Y. A., Alghamdi M. and Alotaiby F., Computer Engineering Department, King Saud University, Riyadh, A. Elmoataz et al. (Eds.): ICISP 2010, LNCS 6134, pp. 122–129, 2010. © Springer-Verlag Berlin Heidelberg, 2010.
[6] Al-zoubaidy L. M., “Efficient Genetic Algorithm for Arabic Handwritten Characters Recognition,” Raf. J. of comp. & Math’s, vol.6, No.2, 2009, received on:29/4/2008,Accepted on :3/9/2008.
[7] Barras C., “Reconnaissance de la Parole Continue : Adaptation du Locuteur et Contrôle Temporel dans les Modèles de Markov Cachés,“ Thesis, Université de Paris IV, 1996.
[8] Ben Nasr M., Talbi M. and Cherif A., “Arabic Speech Recognition by MFCC and Bionic Wavelet Transform using a Multi-Layer Perceptron for Voice Control,” CiiT International Journal of Software Engineering, Vol. 4, No 3, March 2012.
[9] Boite R., Kunt M., “Traitement de la parole,” Presse polytechnique romandes, 1987.
[10] Cheng O., Abdulla W. and Sacic Z., ”Performance Evaluation of Front-end Processing for Speech Recognition Systems,” Electrical and computer Engineering Department School of Engineering, University of Auckland, School of Engineering Report No.621, 2005.
[11] Daqrouq K., Alfaouri M., Alkhateeb A., Khalaf E. and Morfeq A., “Wavelet LPC with Neural Network for Spoken Arabic Digits Recognition System”, British Journal of Applied Science & Technology, 1238-1255, 2014.
[12] Deroo O., “ Modèles Dépendants du contexte et Méthodes de Fusion de données Appliquées à la reconnaissance de la Parole par Modèles Hybrides HMM/MPL “, Thesis, Faculté Polytechnique de Mons, 1998.
[13] El-Ghazi A., Daoui C. and Idrissi N. “Automatic Speech Recognition System Concerning the Maroccan Dialecte (Darija and Tamazight),” International Journal of Engineering Science and Technology (IJEST), ISSN: 0975-5462 Vol. 4 No.03 March 2012.
[14] EL-Mashed Sh. Y., Sharway M. I., Zayed H. H., “Speaker Independent Arabic Speech Recognition Using Support Vector Machine,” ICI-11 Conference and Exhibition on Information technology and Instruction technology, Hungary 2011.
[15] Elshafei M. “Toward an Arabic Text-to-Speech System,” vol. 4B no. 16, pp 565–583, Octobre 1991.
[16] Furui S., “Speaker-Independent Isolated Word Recognition Based on Enphasized Spectral Dynamics,” Procs. IEEE Intl. Conf. on Acoustic, Speech & Signal Processing, pp.1991-1994, Tokyo, Japan 1986.
[17] Ganoun A. and Almerhag I. “Performance Analysis of Spoken Arabic Digits Recognition Techniques,” Journal of Electronic Science and Technology, Vol. 10, No. 2, June 2012.
[18] Génin J., “La parole et son traitement automatique Calliope,“ Annales des Télécommunications, vol. 45, Issue 7-8, pp 457-458, August 1990.
[19] Haykin S., “Neural Networks and Learning Machines”, Prentice Hall, USA, 2009.
[20] Hermansky H., Morgan N., Bayya A. and Kohn Ph., “Rasta-PLP Speech Analysis”, TR-91-069, Decembre 1991.
[21] Kouloughli D. E., “Sur la Structure Interne des Syllabes «lourdes» en Arabe Classique,” vol. 16, numéro 1, pp 129-154, 1986.
[22] Nahar K. M.O, Elshafei M., Al-Khatib W. G. and Al-Muhtaseb H., “Statistical Analysis of Arabic Phonemes for Continuous Speech Recognition,” International Journal of Computer and Information Technology, ISSN: 2279 – 0764 Vol. 01, Issue 02, November 2012.
[23] Rabine L. and Schafer, R., “Digital Processing of Speech signals”, Prentice Hall, 1978.
[24] Satori H., Hiyassat H., Harti M. and Chenfour N., “Investigation Arabic Speech Recognition Using CMU Sphinx System,” The International Arab Journal of Information Technology, Vol. 6, No. 2, April 2009.
[25] Shlens J., “A TUTORIAL ON PRINCIPAL COMPONENT ANALYSIS,” Derivation, Discussion and Singular Value Decomposition, March 2003.
[26] Venkateswarlu R.L.K., Kumari R. V. and Vani Jayasri G., “Speech Recognition Using Radial Basis Function Neural Network”, IEEE, 2011.
[27] Zabidi A., Mansor W., Khuan L. Y., Sahak R. and Rahman F. Y. A., “Mel-Frequency Cepstrum Coefficient Analysis of Infant Cry with Hypothyroidism,” 5th Int. Colloquium on Signal Processing & Its Applications, Kuala Lumpur, Malaysia, 2009.
[28] Zitouni I., Sarikaya R., “Arabic Diacritic Restoration Approach Based on Maximum Entropy Models,” Computer Speech and Language, vol. 23 pp 257–276, july 2009.
Cite This Article
  • APA Style

    Mohamed Hassine, Lotfi Boussaid, Hassani Massouad. (2015). Hybrid Techniques for Arabic Letter Recognition. International Journal of Intelligent Information Systems, 4(1), 27-34. https://doi.org/10.11648/j.ijiis.20150401.14

    Copy | Download

    ACS Style

    Mohamed Hassine; Lotfi Boussaid; Hassani Massouad. Hybrid Techniques for Arabic Letter Recognition. Int. J. Intell. Inf. Syst. 2015, 4(1), 27-34. doi: 10.11648/j.ijiis.20150401.14

    Copy | Download

    AMA Style

    Mohamed Hassine, Lotfi Boussaid, Hassani Massouad. Hybrid Techniques for Arabic Letter Recognition. Int J Intell Inf Syst. 2015;4(1):27-34. doi: 10.11648/j.ijiis.20150401.14

    Copy | Download

  • @article{10.11648/j.ijiis.20150401.14,
      author = {Mohamed Hassine and Lotfi Boussaid and Hassani Massouad},
      title = {Hybrid Techniques for Arabic Letter Recognition},
      journal = {International Journal of Intelligent Information Systems},
      volume = {4},
      number = {1},
      pages = {27-34},
      doi = {10.11648/j.ijiis.20150401.14},
      url = {https://doi.org/10.11648/j.ijiis.20150401.14},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijiis.20150401.14},
      abstract = {In this paper we investigate the use of the feed-forward back propagation neural networks (FFBPNN) for automatic speech recognition of Arabic letters with their four vowels (Fatha, dhamma, Kasra, Soukoun). This investigation will constitute a basically step for the recognition of continuous Speech. Features were extracted from recorded corpus by using a variety of conventional methods such as Linear Predictive Codes (LPC), Perceptual Linear Prediction (PLP), Relative Spectral Perceptual Linear Prediction (RASTA-PLP), Mel Frequency Cepstral Coefficients (MFCC), Continuous Wavelet Transform (CWT), etc. Here, several hybrid methods have been used too. Since the extracted features have large dimensionalities they were reduced by conserving the most discriminatory information with the Principal Component Analysis (PCA) technique. The recognition performance has been improved particularly when we use the PLP method followed by PCA technique.},
     year = {2015}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - Hybrid Techniques for Arabic Letter Recognition
    AU  - Mohamed Hassine
    AU  - Lotfi Boussaid
    AU  - Hassani Massouad
    Y1  - 2015/02/02
    PY  - 2015
    N1  - https://doi.org/10.11648/j.ijiis.20150401.14
    DO  - 10.11648/j.ijiis.20150401.14
    T2  - International Journal of Intelligent Information Systems
    JF  - International Journal of Intelligent Information Systems
    JO  - International Journal of Intelligent Information Systems
    SP  - 27
    EP  - 34
    PB  - Science Publishing Group
    SN  - 2328-7683
    UR  - https://doi.org/10.11648/j.ijiis.20150401.14
    AB  - In this paper we investigate the use of the feed-forward back propagation neural networks (FFBPNN) for automatic speech recognition of Arabic letters with their four vowels (Fatha, dhamma, Kasra, Soukoun). This investigation will constitute a basically step for the recognition of continuous Speech. Features were extracted from recorded corpus by using a variety of conventional methods such as Linear Predictive Codes (LPC), Perceptual Linear Prediction (PLP), Relative Spectral Perceptual Linear Prediction (RASTA-PLP), Mel Frequency Cepstral Coefficients (MFCC), Continuous Wavelet Transform (CWT), etc. Here, several hybrid methods have been used too. Since the extracted features have large dimensionalities they were reduced by conserving the most discriminatory information with the Principal Component Analysis (PCA) technique. The recognition performance has been improved particularly when we use the PLP method followed by PCA technique.
    VL  - 4
    IS  - 1
    ER  - 

    Copy | Download

Author Information
  • LARATSI Lab, ENIM, University of Monastir, Monastir, Tunisia

  • EμE Lab, FSM, University of Monastir, Monastir, Tunisia

  • LARATSI Lab, ENIM, University of Monastir, Monastir, Tunisia

  • Sections