These days, my mind immediately jumps to "what if you plugged all those books into GPT-2!?"
We're also an open source project [https://github.com/internetarchive/openlibrary] and happy to collaborate w/ folks on such projects. I'm personally very inspired by the https://techcrunch.com/2014/07/25/apple-booklamp/ Booklamp project; building a genome for every book and surfacing as much content as we can to help patrons discover citations, quotes, and other useful content which would inform their reading choices and otherwise be completely inaccessible behind a borrow.
If anyone is interested in helping us move the needle on such an effort, please do get in touch and we'd be glad to invite you to Open Library's slack channel.
The ABBYY files are raw output of OCR, but most "items" on the Archive also include the extracted plain text.
It's noisy but generally quite readable. A few heuristics would pass pretty clean copy to GPT-2.