Research project

Automatic Phonetization of Standard Arabic

Phonetization is the process of encoding language sounds using phonetic symbols. It is used in many natural language processing tasks, such as speech processing, speech synthesis, and computer-aided pronunciation assessment. A common phonetization approach is the use of letter-to-sound rules developed by linguists for the transcription from orthography to sound. The target of this project is the improvement of existing transcription rules of standard Arabic and introduction of new rules. A comprehensive algorithm covering the phenomenon of pharyngealization in standard Arabic has already been proposed, and tested on large datasets with an accuracy of almost 100%. Reliable automatic phonetic transcription of standard Arabic has been performed at five levels: phoneme, allophone, syllable, word, and sentence. An encoding, which covers all sounds of standard Arabic, has been proposed. It also allows the automatic generation of pronunciation dictionaries.

Project manager:
Prof. Dr.-Ing. Elmar Nöth

Project participants:
Dipl.-Ing. Fadi Sindran

Keywords:
Standard Arabic; phonetization; transcription rules

Start: 1.1.2011

Contact:
Sindran, Fadi
Phone +49 9131 85 27872, Fax +49 9131 85 27270, E-Mail: fadi.sindran@i5.informatik.uni-erlangen.de

Publications

Sindran, Fadi ; Mualla, Firas ; Haderlein, Tino ; Daqrouq, Khaled ; Nöth, Elmar: Automatic Preparation of Standard Arabic Phonetically Rich Written Corpora with Different Linguistic Units. In: Ekstein, Kamil ; Matousek, Vaclav (Ed.) : Proc. Text, Speech and Dialogue; 20th International Conference, TSD 2017 (Text, Speech and Dialogue; 20th International Conference, TSD 2017 Prague, Czech Republic 28.-31.08.2017). Cham, Switzerland : Springer International Publishing, 2017, pp 201-209. (LNAI Vol. 10415) - ISBN 978-3-319-64205-5
[doi>10.1007/978-3-319-64206-2_23]

Sindran, Fadi ; Mualla, Firas ; Haderlein, Tino ; Daqrouq, Khaled ; Nöth, Elmar: Automatic Phonetization-based Statistical Linguistic Study of Standard Arabic. In: International Journal of Computational Linguistics (IJCL) 7 (2016), No. 2, pp 38-53

Sindran, Fadi ; Mualla, Firas ; Haderlein, Tino ; Daqrouq, Khaled ; Nöth, Elmar: Rule-Based Standard Arabic Phonetization at Phoneme, Allophone, and Syllable Level. In: International Journal of Computational Linguistics (IJCL) 7 (2016), No. 2, pp 23-37

Sindran, Fadi ; Mualla, Firas ; Bobzin, Katharina ; Nöth, Elmar: Automatic Robust Rule-Based Phonetization of Standard Arabic. In: Kral, Pavel ; Matousek, Vaclav (Ed.) : Text, Speech, and Dialogue; 18th International Conference, TSD 2015 (Text, Speech, and Dialogue; 18th International Conference, TSD 2015 Pilsen, Czech Republic 14.-17.09.2015). Berlin : Springer, 2015, pp 442-451. (LNAI Vol. 9302)