This paper investigates the use of hidden Markov models (HMM) for Modern Standard Arabic speech synthesis. HMM-basedspeech synthesis systems require a description of each speech unit with a set of contextual features that specifies phonetic,phonological and linguistic aspects. To apply this method to Arabic language, a study of its particularities was conductedto extract suitable contextual features. Two phenomena are highlighted: vowel quantity and gemination. This work focuseson how to model geminated consonants (resp. long vowels), either considering them as fully-fledged phonemes or as thesame phonemes as their simple (resp. short) counterparts but with a different duration. Four modelling approaches have beenproposed for this purpose. Results of subjective and objective evaluations show that there is no important difference betweendifferentiating modelling units associated to geminated consonants (resp. long vowels) from modelling units associated tosimple consonants (resp. short vowels) and merging them as long as gemination and vowel quantity information is includedin the set of features.