This is where the computer has a great advantage compared to, for example, English learners.
For the computer, grasping all the examples from the dictionary and applying statistics-based analogy to derive pronunciation for the new words is a question of milliseconds.