I will like to know how narrative is done this way with ai voice over

I've observed in the video linked above that the host seems to have control over a button that can halt the voiceover at any moment. It appears as though he's using his natural voice while navigating and demonstrating various elements on the screen. How can I make this type of video?

Can't speak to how successful it will be as I just posted for the first time, but I am planning on moving forward with the AI voice just due to how hard it is to record with kids around.
Just understand that there are major variances in quality even within sites. I spent hours finding a voice that I felt was really well done and, even then, I edit it in audacity afterwards to try to make it a little more human. One other thing I learned doing the first video is that I needed to misspell and use improper punctuation to get some of the sounds and intonations I want.