For those who are interested, I also tried this with Ancient Greek. As expected, it works just as well as with Latin. The fairly large dictionary makes the generated text appear quite “varied.” 
Call: python phonetic_levenshtein.py voyn2greek.txt mapped.txt
Voyn2Greek.zip (Size: 918.43 KB / Downloads: 9)
Note: This version is non-deterministic, which means that each time it runs, the hash function determines a new, best candidate for word matching. This allows you to run the program as many times as you like until you are satisfied with the result. This is most similar to the approach taken by solvers. If you still prefer a deterministic solution, you only need to replace the following file ( after download rename to phonetic_levenshtein.py ):
phonetic_levenshtein.txt (Size: 9.68 KB / Downloads: 4)
Call: python phonetic_levenshtein.py voyn2greek.txt mapped.txt
Voyn2Greek.zip (Size: 918.43 KB / Downloads: 9)
Note: This version is non-deterministic, which means that each time it runs, the hash function determines a new, best candidate for word matching. This allows you to run the program as many times as you like until you are satisfied with the result. This is most similar to the approach taken by solvers. If you still prefer a deterministic solution, you only need to replace the following file ( after download rename to phonetic_levenshtein.py ):
phonetic_levenshtein.txt (Size: 9.68 KB / Downloads: 4)