Advanced Search

Author Topic: Help with Coptic books digitalization & translation  (Read 36 times)

01 November , 2024, 12:47:35 PM
Read 36 times

Offline Admin

  • Administrator

  • *****

  • 1170
    Posts

  • People said thank you: 12

    • Ⲧⲉⲛⲁⲥⲡⲓ
Help with Coptic books digitalization & translation
« on: 01 November , 2024, 12:47:35 PM »
Dear members,

I am making a website for digitalizing & translating old Coptic books.
The website is still in development and I need help from this amazing community to test it and give feedback (feature requests, criticism, suggestions, ...).

There is currently only one book to digitalize for testing purposes but whatever contributions you make to that book, will be migrated to the final version.

Here is the website:

https://ocr.danacbe.com/

For starter you'd want to make an account (no email required).
After that you'd go to any page of the book and you'll be presented with a digitalized copy done by computer along with original image of the page. Your first task is to go through the computer generated text and do make the necessary corrections for it to match the original page.

When two people or more finish correcting the same page (and publish it), it goes through a review process where the website determines the differences between the two contributors and present it to the reviewer to decide which to take and which to reject. Once this step is done the page becomes read only and no more contributions for this page would be accepted.

The corrected text is then fedback to the digitization software (AI) to learn from its mistakes and do better job digitalizing remaining pages.
For that reason digitalziation must exactly match the text of the original page. If original author made spelling mistake, you need to do the same spelling mistake in the digital version.

Also to keep things simple, we ban capital letters. Only use small Coptic letters even if the original is capitalized.

The next step is translation where text are broken into sentences and each sentence can then be translated.
Spelling mistakes corrections and proper formatting are also done in this step.

After that you'd be able to view the page along with its translation in two columns.

The final incomplete step is ability to export the finalized book in various formats (csv, pdf, ePub, ...)

Looking forward to your input


 

Sitemap 1 2 3 4 5 6 7 8 9 10