The phonetic dictionary, which is also known as pronunciation dictionary and lexicon, contains a mapping between the words of the intended language and their corresponding phoneme sequences. It is an essential component for developing an Automatic Speech Recognition (ASR) system. The phonetic dictionary acts as an intermediary component with other ASR components such as the acoustic model and the language model. This research aims to design and develop a phonetic dictionary generator (PDG) for Modern Standard Arabic (MSA) and Arabic disordered speech that can be used for ASR research and development. This rule-based PDG receives Arabic texts and transforms them into their corresponding phoneme sequences using pre-defined rules. Since ASR systems must accommodate different needs of speakers and users including the speech of normal speakers and speakers with speech and articulation disorders, this PDG will be able to produce phoneme sequences for both normal speech and disordered speech using predefined rules too. The input speech from speakers with speech and articulation disorders such as substitution and distortion articulation speech disorders can be different from the speech of normal speakers. In this release of the PDG, the substitution and distortion articulation speech disorders are considered for their importance and frequent recurrence. The PDG is evaluated using an Arabic text that contains 1,623 unique words. The output phonetic dictionary for normal speakers contains 2,473 phoneme sequences, whereas the output phonetic dictionary for speakers with disordered speech contains 62,997 phoneme sequences. This indicates that the output phonetic dictionary for speakers with disordered speech is more comprehensive and contains more possibilities and variations of the same unique word, which would ease the recognition task in ASR systems in a manner that suits different speakers with different pronunciation variations.