You can. There are simple options, that only recognise predefined sentences, that even work on a Raspberry Pi, and at the other end of the spectrum you can host an LLM locally and chat with that if you have the right hardware (Coral isn’t powerful enough for that, you want a GPU with lots of VRAM). Obviously setting this up is more complicated, but there are a lot of options to do it your way.

Depends on which one you have. If you buy their own smart speaker (Voice PE), which is designed to stay entirely local if you have the right hardware and software locally, and even has a hardware switch to temporarily disable the microphone, it’s pretty easy. And of you don’t have all that locally you need a paid subscription to use their cloud a little bit, but they won’t store anything. So still pretty easy.