Most languages are poorly designed.
English is one of the worst. No logical person, starting from a blank slate would design a language like English the way English is constructed. Here’s a straightforward mechanism for simplifying vowel sounds.
Long vowel sounds are singular letters. Short vowel sounds are double letters, or triple letters, or if necessary, quadrupled.
The long “a” sound as in Baker or May should be the singular “a”.
The short “a” sound as in bat, or splat, should be two a’s, “aa”.
The throat “a” sound as in park, or car, should be three a’s, “aaa”.
The long “e” sound as in pee, or flee, should be should be the singular “e”.
The short “e” sound as in fed, or dread, should be the double e, “ee”.
The throat “e” sound as in burn, or tern, should be the triple e, “eee”.
The long “i” sound as in high, or pie, should be the singular “i”.
The short “i” as in drizel or lizard should be double i’s, “ii”.
The pinched short “i” as in ring or drink should be triple i’s, “iii”.
Long “o” as in mope, or dough is “o”.
Short “o” as osprey or problem is “oo”.
Long “u” as in blue, or crew is “u”.
Short “u” as in judge or brother is “uu”.
And that’s it.
If there are other vowel sounds that need to be represented then they can be quadruple a’s or triple u’s, etc.
That is how the language should start out. That is how to design an intelligent language.
And for the consonants? We can probably whittle them down to about 18.
B, D, F, G, H, J, K, L, M, N, P, R, S, T, V, W, Y, Z (missing are C, Q, X)
Any sounds that are more complicated than those should be combinations. “Th” can still be th. Ph? Gone. Gh, like at the end of some words, through, bough, I’m not sure what to do with those.
It might be that we could allow silent consonant combinations at the end of words to allow for homonym differentiation. We should discuss that. There might be some simple rules we could apply that would enhance the language rather than complicate it.