My hypothesis in this regard is that English has specially slow vowels because it’s encoding a lot of information in said vowels. As in: they need to be slower to be distinguishable.

And, when speakers of language A learn language B, they tend to transfer A’s prosody into B. (I believe that this should be uncontroversial as a claim.) That might even get ingrained into a local variety, like Indian English as L1.

So the hard time that people have understanding those Indian English speakers and L2 English speakers from India would be mostly that they don’t get which vowel the Indian speaker is conveying. For example “bit”, “beet”, “bait” sounding almost identical. That goes side-to-side with what you said about “faster than our natural language data speed”, as they’re effectively encoding more info into a certain amount of time than other speakers are able to decode.