Awesome site! This is infinitely more useful than any other pronunciation site I've seen.
Two questions:
I'm guessing the site uses the YouTube API to build a database from video captions, but which videos does it pull from? All of them or a subset? Querying the word "the" yields about 12 million results which seems low to me.
Also, is there any way to prevent the site from modifying my YouTube watch history? I noticed after clicking around a few times and then going back to YouTube's home page that my recommendations had been updated based on the random videos I'd been fed. Clearly this isn't desirable behavior, but I don't know if there's any way around it. For the time being for other users, I recommend using an incognito or private window.