I finally have the Wyoming protocol setup on an external server with RTX 3080ti and it is running very fast. But if I ask it a question, such as “Tell me some facts about Neptune.” It will generate the answer, which can be fairly lengthy which is what I want, but it does not start speaking until the full answer is generated. I see that Piper can stream in wave formt, is it possible to also stream the generated answer to wyoming Satellite instead of waiting for the full answer to be generated. Even if only the first sentence is sent, the rest will be generated by the time the first sentence is spoken. Maybe send the answer sentence by sentence. Thanks for any input.