What process do you use for intensive and extensive reading?

I also prefer physical books, but if the ebook has selectable text, it makes it a cinch to create SRS cards from it, perform lookups, or add digital bookmarks and notes. Admittedly I haven’t used this feature yet, but I am going to try to start.

1 Like

Well, with modern tools and readers you can basically get all ebooks to have selectable text. And on the computer, afaik you can basically always get yomichan to work with whatever book.

1 Like

Well, if the pages were scanned as images and you were relying entirely on OCR, the output would likely require thousands of manual corrections (I speak from experience). In that regard, OCR can be almost as inefficient as just retyping a whole novel, which is what I was getting at by specifying selectable text.

Are you talking about for comics or for books? I haven’t heard of any site that doesn’t give you a handleable ebook format for books.

Ebooks in general. At one point I had quite the eyepatch and peg leg as it were on the internet and had a large collection of questionably obtained ebooks that sadly were scanned as low-quality images. At some point I just started buying books, benefitting the author and being an upstanding internet denizen and all, but mostly it was just to ensure the quality of documents I was receiving.

Even then though, you still occasionally get paid ebooks that are digital images rather than text, which is annoying.

Just saying, if you upgrade your eye patch and peg leg you can find the over 30,000 epubs that a lot of immersion learners have shared, ammased, and organized. But yeah, if you only had poor quality images, I can’t imagine you would have much success.

1 Like

Machine learning / AI goes a long way toward solving this issue. It’s not perfect but gets amazing results in areas where traditional OCR has issues.