Arabic diacritics restoration refers to the process of automatically adding diacritical marks (Harakat) to Arabic text that is usually written without diacritics. The restoration of diacritics is crucial for enhancing the readability, comprehension, and accurate pronunciation of Arabic text, especially for non-native Arabic speakers. But it can also have an impact on the performance of several natural language processing tasks such as Part of speech tagging, Machine translation, text-to-speech, text analysis, information retrieval, etc. This paper aims to describe the different steps followed in the construction of our statistical vocalization tool for Modern Standard Arabic (MSA) text, called MOSHAKIL, which is based on the same principles of statistical machine translation. For the evaluation, we used two corpus, Tashkeela and WikiNews. Our tool demonstrated proficiency in diacritic restoration, achieving diacritic error rates of 06.78% on the Tashkeela dataset and 07.89% on the WikiNews dataset. Similarly, the word error rates were 16.52% for Tashkeela and 18.98% for WikiNews, respectively.