There are videos on youtube of a voice saying a word and its definition for hours, so you play it while you sleep.
The problem with this is you’re going to know a lot of the words. Wouldn’t it be more useful if it generated a mp3 personalized for you, based on words you find difficult? In this hackathon project, we quiz the user to find out what words he doesn’t know, then use a speech library to play the words and definitions as he falls asleep.