Automatic Reading of Text and Numbers in Standard Arabic: The Arphon System
Article Sidebar
Main Article Content
Abstract
This paper deals with speech synthesis from text (TTS ) applied to standard Arabic. The adopted method of synthesis is the concatenation of stored units. The automatic reading concerns texts containing vowelised words and also numbers. The resulting speech is of good quality but has some artifacts for which smoothing procedures are necessary.
Plum Analytics
Artifact Widget
Article Details
In accordance with its open access publishing policy, AL-Lisaniyyat acknowledges and guarantees authors the full and exclusive ownership of copyright and intellectual property rights related to their scholarly contributions.
The publication of an article in the journal does not result in any transfer, assignment, or limitation of these rights. Authors retain full rights over their works, without the requirement to obtain prior written authorization from the journal.
References
[2] S. Lemmetty ‘Review of Speech Synthesis’. MSc. Thesis. Laboratory of Acoustics and Audio Signal Processing, Helsinki. University of Technology, Finland, 1999.
[3] M. Guerti, “Contribution à la synthèse de la parole par diphones en Arabe Standard.’ Thèse de Magistère, Université d’ Al ger, 1983.
[4] A. Hamani, N. Mohallebi, ‘Réalisation d’un lecteur automatique en Arabe Standard sous Word 97", Projet de fin d’études 1998, dirigé par A. Cherif-Zahar, K Ferrat et K. Benbellil.
[5] Y. El-Imam, ‘Speech Synthesis in Arabic Language’, IEEE Transactions of Speech and Signal Processing, 1989.