Towards designing a high intelligibility rule based standard malay text-to-speech synthesis system

• Main Authors: Ahmad, Zakiah Hanim, Khalifa, Othman Omran
• Format: Conference or Workshop Item
• Language: English
• Published: 2008
• Subjects: T Technology (General)
• Online Access: http://irep.iium.edu.my/5916/1/04580574.pdf; http://irep.iium.edu.my/5916/; http://dx.doi.org/10.1109/ICCCE.2008.4580574
• Institution: Universiti Islam Antarabangsa Malaysia

Although text-to-speech (TTS) technology has gained some interest from amateur and professional researchers in developing a Standard Malay (SM) text- to- speech synthesizer, however, up to this day, there is rarely any high intelligible TTS system which is freely accessible to be implemented and introduced to the community of SM speakers. Therefore, identification of the core components required for the development of SM TTS system especially in establishing the NLP module should be carried out intensively. This paper presents a rule-based text- to- speech synthesis system for Standard Malay, named SMaTTS. An intelligible and adequately natural sounding formant-based speech synthesis system with a light and user-friendly Graphical User Interface (GUI) was developed. Result and suggestion for future improvements is discussed. The available Malay TTS synthesizers, the algorithms and speech engine in used, as well as their strong and weak points for each of the system are discussed in this paper. Assessment was made at all possible levels; phoneme, word and sentence level. The overall performance of the system is analyzed using Categorical Estimation (CE) for a comprehensive analysis. Result and suggestion for future improvements is discussed.