I've always wanted to read more then I do, but for whatever reason I have a hard time getting into books because I just have a hard time keeping up with what's going on and visualizing the words on the page. Audiobooks help a little, but it's still not really enough... until now that is.

I recently started reading The Stormlight Archive by Brandon Sanderson, and what I've been doing is after I finish each chapter, I'll put the chapter's text from the e-book into a file and attach it on ChatGPT (now using GPT-4o) and asking the following: "Analyze and summarize the attached text, then generate 3 images depicting different scenes from the text".

I'm pretty amazed at the results. Not only do the summaries help me better understand what I just read, but the generated pictures help me visualize the scenes better then I ever thought possible. Here's one picture it generated that I'm really impressed with:

https://preview.redd.it/2ul6tiqt1z0d1.jpg?width=1792&format=pjpg&auto=webp&s=c7272d78f0b26b720cb8e023f32442772a50441a

This is a scene from a chapter showing the aftermath of a battle, where Kaladin (the main protagonist) is a slave who's bridge crew was attacked leaving him clinging to life. It even accurately portrayed Sylphrena, the tinkerbell-like spren that's befriended Kaladin.

This really is amazing, at least to me. Hopefully in the future it can even generate audio or even video based text.