A System of Automatic Speech Recognition based on the Technique of Temporal Retiming

We report in this paper the procedure of a system of automatic speech recognition based on techniques of the dynamic programming. The technique of temporal retiming is a technique used to synchronize between two forms to compare. We will see how this technique is adapted to the field of the automatic speech recognition. We will expose, in a first place, the theory of the function of retiming which is used to compare and to adjust an unknown form with a whole of forms of reference constituting the vocabulary of the application. Then we will give, in the second place, the various algorithms necessary to their implementation on machine. The algorithms which we will present were tested on part of the corpus of words in Arab language Arabdic-10 [4] and gave whole satisfaction. These algorithms are effective insofar as we apply them to the small ones or average vocabularies.

Authors:

• Samir Abdelhamid
• Noureddine Bouguechal

Keywords:

• Continuous speech recognition
• temporal retiming
• phonetic decoding
• algorithms
• vocal signal
• dynamic programming.

References:

[1] J. P. Haton, D. Fohr et M. Djoudi: Un système expert pour le décodage
acoustico-phonétique pour l-Arabe standard. Conférence Maghrébine,
Septembre 1989.
[2] Y. Belkaid: Les voyelles de l-Arabe littéraire moderne. Analyse
spectrographique Rapport N┬░ 16, travaux de l-institut de phonétique de
Strasbourg, 1984.
[3] O. Deroo, C. Ris: Hybrid HMM/ANN Systems speaker independent
continuous speech recognition in French Travaux de l-école
Polytechnique de MONS Belgique, 2000.
[4] S. Abdelhamid: Contributions ├á l-étude et ├á la réalisation d-une machine
├á dicter en Fran├ºais. Thèse de Magister de l-institut d-informatique de
l-université de Batna, Algérie, 1994.
[5] M. Guerti: Contribution ├á la synthèse de la parole en Arabe standard.
Actes des 16ème journées d-études sur la parole. Hammamet, Tunisie
1987.
[6] Benhamouda: Morphologie et syntaxe de la langue Arabe. Nationale
Edition, 1983.
[7] N. Carbonell, J. P. Haton, D. Fohr Aphodex, design and implementation
of an acoustic-phonetic decoding expert system. IEEE International
conference on Acoustics, speech and signal processing, 1986.
[8] V. Barreaud : Reconnaissance automatique de la parole continue:
compensation des bruits par transformation de la parole. Thèse de
l-université de Nancy1, 2004.
[9] S. Stuker :Automatic Generation of Pronunciation Dictionaries For New,
Unseen Languages by Voting Among Phoneme Recognizers in Nine
Different Languages, Master thesis, Carnegie Mellon University,
Pittsburgh, PA, USA, April, 2002.
[10] D. Vaufreydaz, M. Akbar, J. Caelen : Environnement Multimédia pour
l'Acquisition et la gestion de corpus Parole, JEP'98, pp. 175-178,
Martigny, Switzerland, June 1998.
[11] H-F. Silverman, D-P. Morgan: The application of dynamic
programming to connected speech recognition, IEEE ASSP magazine,
vol.7, pp.6-25, 1990.
[12] L. R. Rabiner : A Tutorial on Hidden Markov Models and Selected
Applications in Speech Recognition, L.R. Rabiner, Proceedings of the
IEEE, vol 77, No 2, 1989.
[13] N. Carbonell, J.P Haton, F. Lonchamp, JM. Pierrel : Élaboration
expérimentale d'indices prosodiques pour la reconnaissance; application
├á l'analyse syntaxico-sémantique dans le système MYRTILLE II",
Séminaire Prosodie et Reconnaissance, Aix-en-Provence, 1982.
[14] J. M. Pierrel : Utilisation des contraintes linguistiques en compréhension
de parole continue dans le système Myrtille II. TSI, Vol 1, N┬░ 5, 1982,
pp. 403-421.