Very cool!
Though... I'm wondering if maybe you're taking the transliterating a little too far? I haven't actually run your code, but from looking over the spoiler blocks, it seems like you might be doing a lot of unnecessary replacements, particularly when it comes to vowels.
There was a quote from Peter Ahlstrom on the old forum: "The person who translated these books into English treated certain art pages different ways in order to tailor it to the English-speaking audience. The Navani pages are meant to give a flavor for what the writing looks like, yet still be something readers can figure out and understand."
Unless there is other information outside of the books that I've missed (which is entirely possible -- I haven't kept up with the forums much), I don't think it's really possible to transliterate English into *actual Alethi*, only the pseudo-Alethi that's used in the WOK artwork. So, if your goal is to produce Alethi script consistent with what we have in WOK, I don't really think you need to convert English into full phonetic notation before applying the Alethi script font. Apart from the trickyness with c > s/k, I think all you really need to do are some basic character-level replacements.
The examples we have of pseudo-Alethi (the excerpts from Navani's Notebook on p. 762 & 856)follow English orthography pretty straightforwardly, as I recall -- apart from a few specific exceptions at the character level:
Th > /θ/
CH > /tʃ/
C > /k/ or /s/
W > /ʌ/
X > /ks/
I forget if there's anything canon regarding Qu, but I think it should realize as either /k/ or /kʌ/
For example, the device on p. 762 is labelled just as it would be spelled in English ("Pain Knife"), not as it would be pronounced (/pān nīf/). Furthermore, the English word "Joy" (/dʒoi/) is written using distinct Alethi characters for "J" and "Y" even though that combination of letters would pronounced in modern Alethi the same as "Yoy" (/joi/).
But regardless, I'm impressed with what you folks have done!