This is not an easy problem. The relevant Unicode Annex #15 explains: > The visual appearances of the compatibility equivalent forms typically constitute a sub

This is not an easy problem. The relevant Unicode Annex #15 explains: > The visual appearances of the compatibility equivalent forms typically constitute a subset of the expected range of visual appearances of the character (or sequence of characters) they are equivalent to. However, these variant forms may represent a visual distinction that is significant in some textual contexts, but not in others. As a result, greater care is required to determine when use of a compatibility equivalent is appropriate. If the visual distinction is stylistic, then markup or styling could be used to represent the formatting information. However, some characters with compatibility decompositions are used in mathematical notation to represent a distinction of a semantic nature; replacing the use of distinct character codes by formatting in such contexts may cause problems. In the context of a mathematical text where the semantic difference is relevant, you wouldn’t want NFK[CD] to be blindly applied. I agree, however, that screen readers shouldn’t silently omit those characters mid-sentence. At the very least, they should say something like “five unsupported characters omitted”.