We've created a library named "DSAraby" that aims to transliterate text which write a word using the closest corresponding letters of a different alphabet or language. The algorithm gives the possible words in Arabic based on a given word in Latin by mapping Latin letters to Arabic ones, then takes the most frequent word existing in a corpus.
from dsaraby import DSAraby
ds = DSAraby()
w = ds.transliterate('chokran likom kamliin')
print(w)