The images of spectrogram analysis between the real and fake voices seemed to be distinguishable by the human eye. Can a image model be trained to detect fake voice spectrograms based on pitch and tone choppiness?
I find the site decently fast, definitely ugly but then again I don't want it to get a reddit-style redesign either. The information density is ok right now, and I'm actually impressed by the wide range of functionalities they have, related to reviewing books and updating your progress.
Working closely with people (friends preferably) who have higher "discipline" than you, and seeing how they grind their way through a problem is instructive sometimes.