Hello,
More of a general consideration, since I’m guessing the only way of resolving this is on a sentence by sentence basis, but: in languages where verb forms change depending on the speaker’s grammatical gender, it can be slightly confusing that the text to speech voice and the grammar don’t match up. Thoughts? E.g. in Russian, currently voices that use the verb forms that female speakers would use are currently pronounced by a male voice or vice versa, since the sentences include both sets of verb forms but the app always stays with either a male or a female voice.