Asterisk is flexible, so with the right tool I do not see why you couldn’t. There are plenty of examples of people doing this with TTS and ASR.
To not directly answer your question I haven’t read anything on anyone doing specifically what you are asking for, but it might be worth it to ask the vendor if they are aware of an Asterisk use-case involving their product? It seems like there is a free trial and paid support. I would imagine they would be one of your best points of reference to determine if what you are looking for is feasible.