With all of the Google, Amazon, and Apple Homepod news from CES, it is clear that there is money to be made in mining voice for data.
This inspired me to go check up on the Mycroft
project, which is doing fine.
I appreciated the January 11th blog post
that explained how Speech to Text works. It looks like they aggregate all queries into a system without any personal identifiers (like DuckDuckGo). STT engine can be changed as well.
The Kickstarter for the Mark II voice assistant
was funded within six hours of launch. I should drop it in a VM, and see what it can do. I could drive my wife nuts by creating a module loaded with dad jokes.