Abstract—A statistical parametric speech synthesis system based on hidden Markov models (HMMs) has grown in popularity over the last few years. In this approach the system simultaneously models spectrum, excitation, and duration of speech using context-dependent HMMs and generates speech waveforms from the HMMs themselves. In this paper, the HMM-based speech synthesis system is applied to Arabic language using low size unsegmented speech training database. This technique shows that the resulting HMM set has the advantage of being small (can be less than 1MB) which is very important for communication applications. The basic contribution in this paper is to justify both the HMM parameters and the speech features to be suitable for using small speech database to get the highest quality. The motivation of this work is the starvation of the Arabic speech database. Experiments show that using Mel-cepstral coefficients as spectral parameters of speech waveforms for training gives better results than using LPC or PARCOR coefficients. Also, investigation tests show that increasing the context-dependent models length and the number of Gaussian Mixtures with this relatively small size training data has the disadvantage of poor generalization of HMMs that leads to perceivable discontinuities and clicks in the synthesized speech.

Index Terms—Speech synthesis, Arabic language, Hidden Markov Modeling, and Mel-Cepstral coefficients.

M. S. Barakat, He is now with the Department of Computer Science, Modern Academy, Maadi, Cairo, Egypt (e-mail: msamirhb@hotmail.com).
M. E. GadAlla, was with Military Technical Collage, Cairo, Egypt. He is now with the Department of Computer Science, Modern Academy, Maadi, Cairo, Egypt (e-mail: mahmoud_mtc@yahoo.com ).

Cite: Mohamed Samir Barakat and Mahmoud El-Said Gadallah, "The Effect of Speech Features and HMM Parameters on the Quality of HMM Based Arabic  Synthesis  System," International  Journal  of  Computer  and
Electrical Engineering vol. 2, no. 2, pp. 235-242, 2010.