That makes sense. TTS is commonly set up as a daemon, so while you can run an external process, you can't just monitor when the voice is supposed to have ended (unless there's a command for that as well...). But in that case I'd like to be able to disable the robot voice altogether.