Is it possible to use phonetic analysis algorithms to determine transliteration?

Question

Audio

I read on Habré about Soudex and Metaphone and now I came up with a task for myself:
Having a text at the input using phonetic algorithms, determine whether it is an English-Russian tranlit or not.
Actually, I immediately ran into the problem that it is impossible to compare the results of the English and Russian results of the algorithms.
If someone solved a similar problem, tell me in which direction to look?

Answer 1

Here is an example of this.
Not bad done: "iPhone", "ipod", "nexus red" - looking for!