Open access, open source, open to all

Devoted to discussing any and all manuscripts written in Arabic script and derivatives.

OCR

1 voice, 0 replies

Viewing 0 reply threads

Author

Posts
- 30 December 2022 at 11:35 am EST #64059
  
  Jake Benson
  Participant
  @jakebenson14850
  
  Terrific news for non-Latin script specialists: Archive.org will take advantage of new advancements in OCR to reprocess printed multi-script manuscript catalogues, which currently misinterpret non-Latin scripts as gibberish. They just reprocessed one volume of Rieu’s British Museum catalogue https://archive.org/details/gri_33125008986313 and Sachau and Ethé’s for the Bodleian https://archive.org/details/catalogueofpersi01bodluoft as well. Should any of you encounter specific books that lack functional non-Latin script, write to support@archivesupport.zendesk.com, send them the URL along with any 3-letter Library of Congress language codes https://www.loc.gov/marc/languages/language_name.html so they can drop that in and reprocess the files for multiple languages. Now, if only we could somehow persuade Google to undertake the same initiative!
Author

Posts

Viewing 0 reply threads

You must be logged in to reply to this topic.

User Groups

@

Not recently active