Autotune algorithms have some of that although maybe not to the degree you’re talking about.
Something about hamming windows is coming to mind.
I saw a fascinating resource on that i played with a few years ago — lemme find it:
I found the minimum euclidean distance of sound but that’s not what you’re talking about either. You’re talking about shifting it from an equal temperament to natural intonation.
Anyway, these two were the scope of my sound experiments that were vaguely similar but not the same as what you’re talking about.