Virtual Assistant

[NavigSB]

How feasible is this? Libraries? How to use STM32 as slave for Pi? Is that even a good idea? I think it might be the only way...

Research doc.

[NavigSB]

This feature is actually even more do-able than I had first expected. The tools are just ready to go on GitHub and ready to be tinkered with. The two libraries that I really liked were Rhasspy for voice recognition and other voice assistant features, and EspNet for text-to-speech. The only things I'm worried about is whether either is too resource-intensive (they probably aren't, but it's possible), and that it might be slightly hard to connect the two (this also seems unlikely).