Investigation Arabic Speech Recognition

From signal to its interpretation


H. Satori1, 2*, N. Chenfour1, 2, M. Harti1, 2

1 UFR Informatique et Nouvelles Technologies d'Information et de Communication B.P. 1796, Dhar Mehraz Fès morocco

2 Département de Mathématiques et Informatique, Faculté des Sciences, B.P. 1796, Dhar Mehraz Fès, Morocco

* Corresponding author. E-mail:

Received: 03 January 2007; revised version accepted: 12 March 2008



     In this paper we present an adaptation of CMUSphinx-4 speech recognition system to Arabic environment. Our approach for modeling Arabic sound consisted of generated and trained acoustic and language models with Arabic data. Different parameters in the system were adjusted to fine tuning the system for Arabic, some are training and other are decoding.


Keywords : Speech recognition; Acoustic model; Arabic language; HMMs; CMUSphinx-4; Artificial intelligence.

© 2015