Diacritic marks are part of a system of notation that show with detail how words are pronounced. A system of broad transcription represents the building blocks or phonemes for a word. For example, in the International Phonetic Alphabet (IPA), the word pot could be broadly transcribed as /pɑt/. In that alphabet, these letters represent the phonemes /ppppppppp/, /ɑäahɑäahɑäah/, and /ttttttttt/. (Here the vowel symbol represents phonemes as in hot or hop, not the short /æăaæăaæăa/ as in hat.) Diacritics may be added to this transcription to add detailed information about its pronunciation.
In narrow transcription more detail is added about pronunciation by using diacritic marks. For example, The same word may be written in IPA as [pʰXXXXpʰXXXXpʰXXXXɑäahɑäahɑäahttttttttt]. Here you see two additional symbols, or diacritics, have been added. The superscript h (ʰ) indicates that the first consonant is likely to be pronounced with a puff of air. The upper right corner diacritic on the /t/ [ttttttttt] indicates that the /t/ is likely unreleased. In other words, the tongue will raise to the gummy ridge behind the upper front teeth and hold in place.
In general, Cued Speech represents broad transcription, not narrow transcription. We cue only the general building blocks of language assembled according to the rules of English. Narrow transcription deals with the details of speech production for a word in that context at the moment of being uttered. Cueing does not have cues to represent all of the possible pronunciations for each phoneme.