MobileRead Forums - View Single Post

Neog2 · 05-16-2025, 05:30 PM

Well I know this isn't a solution for everyone but as I only purchase a couple books a week my solution to convert books to epub works 100% just the process is tedious and takes a bit of time. But instead of trying to crack the DRM i just bypass. I'm hoping that someone can make a script or a A.I. Agent to automate the process.

Windows 11. Current version of the Snipping Tool has OCR capability built in.
Amazon Cloud Reader, or the latest version of Kindle for PC both can be used.

1. Kindle for PC I can get the font size down to where it will show 3 pages of text per page turn. and still be able to correctly be read by Windows "Text Actions" >"Copy all text" feature and then place in a word document.

So the Book i did today had 190 screenshots including Title Page, and 2 Pages of Table of Contents screenshots. File names are the Beginning Page number. So for today's Book it went like this.

1. Title Page
2. Table of Contents 1
3. Table of Contents 2
4. Page 013 Chapter 1
5. Page 016
6. Page 020 Chapter 2
7. Page 023
8. Page 026 Chapter 3

Then I extracted Text into a Narrow Margin No Space Word Document. For my use case Word Documents, PDF's and Epub's all work. But its not hard to convert a doc to epub.

I have also tested this with Open A.I.'s Operator (A.I. Agent ) using a Google drive folder and a free OCR website. It does work but you need to be very specific about the prompts and it needs to download each picture individually then upload each one to the free website to extract the text but it will take all the text and put in a doc for you. Unsure why Operator doesn't have the capability of ChatGpt when it comes to directly extracting text from a Picture.

ChatGPT also works in testing to extract text from pictures but you are limited to 10 pics at a time and its considerably slower than just doing Snipping Tool, copy all text, then paste directly into a word document.

And down the line even if they prevent you from taking screenshots with Snipping Tool in windows you can always use something like OBS screen capture then extract text using the Snipping Tool directly on OBS. Or even a step further if you have two pc's and a capture card do the exact same process mirroring image onto a second pc then use your OCR program of choice.

Obviously this process is Tedious and works vastly better with books that are basically only text. If your book has tons of tables or charts like in LitRPG/GameLit type of books then you will have to do some clean up manually every so often.

Just posting as a work around and maybe the script kiddies or the people that are currently gathering to make useful A.I. Agents can ultilize their skills and just bypass this whole DRM crap for purchased books.

05-16-2025, 05:30 PM	#2927
Neog2 Junior Member Posts: 1 Karma: 10512 Join Date: May 2025 Device: Kindle	Well I know this isn't a solution for everyone but as I only purchase a couple books a week my solution to convert books to epub works 100% just the process is tedious and takes a bit of time. But instead of trying to crack the DRM i just bypass. I'm hoping that someone can make a script or a A.I. Agent to automate the process. Windows 11. Current version of the Snipping Tool has OCR capability built in. Amazon Cloud Reader, or the latest version of Kindle for PC both can be used. 1. Kindle for PC I can get the font size down to where it will show 3 pages of text per page turn. and still be able to correctly be read by Windows "Text Actions" >"Copy all text" feature and then place in a word document. So the Book i did today had 190 screenshots including Title Page, and 2 Pages of Table of Contents screenshots. File names are the Beginning Page number. So for today's Book it went like this. 1. Title Page 2. Table of Contents 1 3. Table of Contents 2 4. Page 013 Chapter 1 5. Page 016 6. Page 020 Chapter 2 7. Page 023 8. Page 026 Chapter 3 Then I extracted Text into a Narrow Margin No Space Word Document. For my use case Word Documents, PDF's and Epub's all work. But its not hard to convert a doc to epub. I have also tested this with Open A.I.'s Operator (A.I. Agent ) using a Google drive folder and a free OCR website. It does work but you need to be very specific about the prompts and it needs to download each picture individually then upload each one to the free website to extract the text but it will take all the text and put in a doc for you. Unsure why Operator doesn't have the capability of ChatGpt when it comes to directly extracting text from a Picture. ChatGPT also works in testing to extract text from pictures but you are limited to 10 pics at a time and its considerably slower than just doing Snipping Tool, copy all text, then paste directly into a word document. And down the line even if they prevent you from taking screenshots with Snipping Tool in windows you can always use something like OBS screen capture then extract text using the Snipping Tool directly on OBS. Or even a step further if you have two pc's and a capture card do the exact same process mirroring image onto a second pc then use your OCR program of choice. Obviously this process is Tedious and works vastly better with books that are basically only text. If your book has tons of tables or charts like in LitRPG/GameLit type of books then you will have to do some clean up manually every so often. Just posting as a work around and maybe the script kiddies or the people that are currently gathering to make useful A.I. Agents can ultilize their skills and just bypass this whole DRM crap for purchased books.