The greatest difficulty in reading Arabic script for nonnatives is the absence of short vowels. The correlation of 28 consonants with sounds poses no great difficulty. In Arabic, there are six vowel phonemes which are voiced only by three letters with help of some relevant diacritical marks (?arakat). As the bulk of Arabic publications is written without such marks, foreign readers are at a loss to read the written work intelligibly unless they are familiar with the grammar of the language. Especially practicing Muslims who are required to read the Islamic Scripture Qur'an even without a deep knowledge of its language are unable to produce acceptable articulation. In order to meet the needs of such nonnative believers and the learners of the Arabic language, many attempts have been made to transliterate (romanize) the texts replacing letters with their close equivalents in the Latin alphabet by some modifications. Although this method has helped somewhat in decoding the consonants, the issue of vowel representation has always been an unresolved issue. The IPA advocates suggested corresponding vowels from its files and that seems to have solved at least the segmental part of the problem. However, knowing that the overall speech intelligibility lies more with prosody, i.e. proper rhythm and intonation, the author believes that markings relating to such features should also be added to the transcription. This article presents an innovative approach to Arabic transcription using a pattern on a popular scriptural text, i.e. the "Surah Yousuf" the first time ever in literature. First the grapheme-morpheme transcription of the text with "i'rab" has been used according to ALA-LC transcription. After the meaning chunks have been further separated by pause signs, word stress levels and linking cases were also marked. Thus it is hoped that a more intelligible Arabic IPA transcription is reached for those interested in reading and teaching to read Arabic script clearly for personal or scholarly purposes.