Title: Adapting eSpeak to Arabic language: converting Arabic text to speech language using eSpeak

Authors: Taha Zerrouki; Mohammed M. Abu Shquier; Amar Balla; Nabila Bousbia; Imededdine Sakraoui; Fateh Boudardara

Addresses: Laboratoire des Méthodes de Conception des Systèmes, Ecole nationale Supérieure d'Informatique (ESI), BP 68M, 16309, Oued-Smar, Algiers, Algeria ' Department of Computer Science, Faculty of Computer Science and IT, Jerash University, Jerash, Jordan ' Laboratoire des Méthodes de Conception des Systèmes, Ecole nationale Supérieure d'Informatique (ESI), BP 68M, 16309, Oued-Smar, Algiers, Algeria ' Laboratoire des Méthodes de Conception des Systèmes, Ecole nationale Supérieure d'Informatique (ESI), BP 68M, 16309, Oued-Smar, Algiers, Algeria ' Laboratoire des Méthodes de Conception des Systèmes, Ecole nationale Supérieure d'Informatique (ESI), BP 68M, 16309, Oued-Smar, Algiers, Algeria ' Laboratoire des Méthodes de Conception des Systèmes, Ecole nationale Supérieure d'Informatique (ESI), BP 68M, 16309, Oued-Smar, Algiers, Algeria

Abstract: Text to speech (TTS) is a crucial tool needed in many domains, mainly for visually impaired users. The availability of TTS open sources improves access to computers and gives more valuable applications. eSpeak provides support for several languages. It is a tool that provides rules and phoneme files for more than 50 languages, besides, eSpeak is a light, fast, low memory consumption and used in multi-platforms. In this paper, we have explored the possibility to adapt the existing text to speech converters into Arabic language in eSpeak. We attempt to define new text to speech conversion rules, adapting existed phonemes and adding missing phonemes for Arabic under eSpeak. The contributions are quite significant; however, the software's developers will be able to integrated these enhancements within the new version, so that users who have problems with visual impairments or children with special needs will utilise this development of eSpeak. The availability of such support, open new fields to use Arabic in TTS environment, especially for blind persons.

Keywords: text to speech; TTS; eSpeak; Mbrola; Arabic; open source.

DOI: 10.1504/IJRIS.2019.098056

International Journal of Reasoning-based Intelligent Systems, 2019 Vol.11 No.1, pp.76 - 89

Received: 23 Dec 2016
Accepted: 11 Jan 2018
Published online: 01 Mar 2019 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article