I'm okay with it, despite my first exposure being that joke in the car and I wasn't expecting it. That's what I get for not paying attention to the changelog then. But while I wasn't sure at first, I think the current way you're doing it is perfectly fine. I'm sure it'll take some time to get it all down anyway, but a suggestion is that you want to do most of the voice recording close together. Quality, time, etc. That way you won't get different results for later lines. And so long as there's an option to turn it on/off then I can't say anything else about it. Story should be built up first, of course. That's important.