I am non-native in English, so this question may be a meaningless one or even a silly one. Why do we need a stress on one or more letters in a word? Indeed, a native person can read a word containing stress; even I omit duplicated letters. Is this just a heritage from the historical times, or not?
In general, it appears that "rhythm" of speech in one form or another improves intelligibility(*). Syllable "stress"— making particular syllables locally prominent in some way compared to other syllables — is one component of rhythm. It may not be strictly a necessary feature of language, but it appears to be a perceptually useful feature. It is possible that all languages use stress in one way or another (see Hirst & Di Cristo, "Intonation Systems", for a survey of how a selection of different languages from different language families use stress and intonation generally).
Now, although stress per se is an apparently universal feature of language, lexically contrastive stress as in English — i.e. where the stress position can be determined by the identity of the word in question and the difference between two words can be determined by stress pattern — is by no means universal. In many languages, such as French, which syllables carry which type of stress is essentially determined at phrase level and not dependent on the identities of particular words (though in such a language, you can still get certain classes of words that cannot carry stress vs others which do).
So in summary: rhythm/intonation is perceptually important to languages and stress patterns form part of what we perceive as "rhythm". But the precise system of stress used varies from language to language and there is little necessity for English to have the particular stress system that it does.
(*) See e.g. Tajima et al (1997), "Effects of temporal correction on intelligibility of foreign-accented English", Journal of Phonetics 25:1–24. Even with speech in which, say, the actual quality of vowels is quite far from that of a native speaker, by correcting specifically the timing of those vowels (and other segments in the speech), these researchers found that speech becomes more intelligible despite the fact that the speaker is actually pronouncing 'the wrong sound' as it were.
When an English speaker de-stresses a syllable, he or she will give a little less time to the pronunciation of the syllable, and move the vowel closer to a neutral vowel (schwa or ə).
Nouns and related verbs sometimes have different stresses to indicate that they are different parts of speech.
I suspect that many native readers would have to read that sentence twice to get the pronunciation of the repeated words correct. The would be pronounced with stresses as:
A native speaker would know that the first "record" is an adjective and the second "record" is a verb.
To answer your question: English has evolved to have a more consistent orthography, while sacrificing a less consistent pronunciation. So words like
have a common orthography, even though the second "c" in the first word sounds like "k" and the second "c" in the second word sounds like "s". Also, the stress in the first word is on the second syllable, but in the second word is on the third syllable.
English orthography doesn't mark stress. This makes the job easier for the reader and writer, but harder for the speaker and listener.