Amritas writes a nice post, in which he talks about the word 'shalom', in the context of Semitic roots and patterns. He also links to this table in JPG form. For me, at least, the table is quite blurry and hard for me to read, so I have reproduced it below. In the process, I have turned it on its side, and replaced some of the original graphs in order to make it more amenable to the blog and HTML format. I also added a column for my own Hebrew transcription, that I use on this blog, which is meant to represent a superset of modern Hebrew pronunciations.
Original Notes: (1) In Akkadian the consonants gh, `, h, ', h, and usually y were lost, although there is evidence that they were present in the oldest stages of the language. (2) In Hebrew and Aramaic the non-emphatic stops b, p, d, t, g, k become fricatives (pronounced v, f, dh, th, gh, kh, respectively) after vowels unless they are doubled; *w at the beginning of words became y.
My Notes: (1) Underlined letters and q, except for h, are "emphatic". The original pronunciation of emphatic consonants is unclear, but in modern Hebrew and Arabic they are velarized, or uvularized (the back of the tongue is raised during pronunciation, in the case of q, this means that it is pushed back to a uvular stop). (2) In Modern Hebrew the process of stops regularly becoming fricatives, described above, has reversed itself for dh, th, and gh, which have reverted to d, t, and g respectively. Modern Hebrew spelling reflects the sound system in the Hebrew column of this table, which does not correspond one-for-one with my transcriptions. (3) By and large, Semitic languages are quite closely related, more comparable to the Germanic language family than the Indo-European family. Armed with this table, it is quite easy for amateurs to figure out cognates. For example, when I hear an Arabic word, I can usually come up with a Hebrew cognate.
|Proto-Semitic||Akkadian||Ethiopic||Arabic||Aramaic||Hebrew||my Hebrew transcription|
|labial stops||b||b||b||b||b||b||b, v|