This is more generally the concept of Orthographic depth [1]
A recent study attempted to categorize languages and found English and French to be among the most opaque, Finnish and Korean to be the most transparent, and German in the middle. [2]
A recent study attempted to categorize languages and found English and French to be among the most opaque, Finnish and Korean to be the most transparent, and German in the middle. [2]
Now the orthography is hard mainly due to: (1) change in the language not reflected in the writing, (2) etymological speeling choices made on standardization and (3) latent phonological information.
Instances of (1) include graphemes likes "au" [o].
Example of (2) is "pied" with final letter here to refer to Greek "podo".
Point (3) is more complex. Take "parent" were final t is mute. Its presence makes is clear it’s related to "parental". Some with pairs like "grand" / "grande" were in the feminine form the final d is pronounced.
Edit: Tibetan alphabet is notoriously difficult, and Thai alphabet is quite hard too, so the problem exists elsewhere as well even when staying in the realm of alphabets.
The French result is interesting. It was fine - not great, not terrible - on the readability score, and by far the worst on writability. (The 2nd worst, English, got 3x the writability score.) That, combined with the fact that this metric was based on how much trouble a neural net had goes a long way toward explaining why my phone can do text-to-speech in French just fine, but can't take dictation to save its life.
They're also used for grammar practice. In that case, the challenge is that there might be many written forms of the sentence that would be pronounced the same as what was said, but only one is grammatically correct.