It could be useful to provide alternative sentence text from which audio is generated. One example where this is useful are common abbreviations: when reading the student would want to see the text like it is usually written, with abbreviation, but when listening the abbreviation would be expanded to the text it stands for. This way we’d be getting used to say the actual words in stead of abbreviations.
Another use is replacing punctuation: the current audio generator treats many symbols as sentence ends (e.g. ! or …) but often they indicate only a phrase end within a sentence. It would be useful to replace them with commas or semicomas in a sentence send to audio generator, so the pause doesn’t break the sentence flow.
It could also be useful in some other, less common, situations. For example, some texts contain comments in parentheses, which I would like to see when reading, but don’t want to hear when listening (e.g. years of events in a text about history, when they aren’t actually a part of the sentence, I find these parentheses distracting when listening, but don’t want to lose them entirely).
Yet another use is even more marginal, but the generator seems to have difficulties with some spellings of older Turkish words and again I would like to see them as in original, but to hear them with correct pronunciation, so I would want to rewrite those words in the alternative text.
Not every sentence needs to have this alternative version, moest wouldn’t, but it would be nice to have the option to override what is used to generate audio.