It can be fine tuned for device related actions. In other words, with all the capabilities of your device applications or services, the small model can virtually have the same capabilities. It can always dispatch a user request in way of “natural language” to those applications, and orchestrate the applications. It can dispatch user requests beyond the device capabilities to a cloud model. This is powerful since it changes how you interact with your devices.
It is still very useful to standardize the pronunciations, since people with different dialects had to meet especially those officials in government. There was “yayan” for this purpose.
I've also tried a few Android LLM apps, all running more than 30min.
Current LLM models are not running constantly on the phones to drain your battery. They just run when responding to a prompt. It by no means consumes more battery than a heavy game.
The labor replacement is happening in the whole human history, but I am not worried, not because I don't believe LLM's capabilities in 10-15 years, but because I believe I can always find some areas where LLM is not good at.