All productsDigital transformation and operations
Archive OCR
OCR for Vietnamese history books, with quality you can measure
Turns scanned Quốc ngữ history books into searchable text and scores every page, so reviewers only open the pages that need a second look.
What it does
-
Built for Quốc ngữ
Recognises Vietnamese, with all its diacritics, in scanned history books.
-
A score for every page
Word confidence, the share of low-confidence words and the share of valid Vietnamese words, so weak pages stand out.
-
Mixed-script detection
Flags pages where Hán characters appear among the Vietnamese text.
-
Measured against people
Character and word error rates against pages a person has checked.
-
Throughput you can plan with
Seconds and words per page, recorded for every run.
Want something like this for your business?
Tell us about the process or product you have in mind. We reply by email with questions or a time to talk.
Talk to us