News for hackers. Stuff that matters.
Re:Mistral OCR by raffraffraff on Friday March 07, 2025 @03:33AM (#43288305)
It feels like, after the OCR step there should be language and subject matter detection, with a final sweep with a spelling / grammar checker that has the right "dictionary" selected. (That, right there, is my naivety on the subject, but I would have thought that the type of problem you're describing isn't OCR but classical spelling and grammar checking?)
Re:Mistral OCR by layer8 on Friday March 07, 2025 @08:02AM (#43289822)
It’s OCR in that the wrong characters are being recognized. This is not about fixing spelling or punctuation mistakes present in the source image, it’s that errors are being introduced due to a lack of accuracy of this OCR with regard to punctuation and typography. The punctuation errors are not different in principle from the case of producing a misspelled word that wasn’t misspelled in the original.

A subsequent cleanup pass that fixes grammar/spelling errors wouldn’t be appropriate when the goal is to faithfully reproduce the original text.

Re:Mistral OCR by raffraffraff on Friday March 07, 2025 @05:02PM (#43295245)
> A subsequent cleanup pass that fixes grammar/spelling errors, as you propose, wouldn’t be appropriate when the goal is to faithfully reproduce the original text

I suppose that depends on why it's wrong. Did the model accurately read a real typo in the image or did it incorrectly decipher a character? If a spelling & grammar pass fixes the latter, isn't it valid?

Re:Mistral OCR by pbhjpbhj on Friday March 07, 2025 @10:39AM (#43291028)
Not unrelated - OneNote 'copy text from image' has started producing lots of incorrect OCR results, but they're all non-words.

For example, from a clear image of a printed page (in a standard font), it will give me 'cornprising' instead of 'comprising'; 'niatter' instead of 'matter'. Excepting the spell-check underline they'd be hard to spot as with relatively tight kerning all the errors look like the originals.

I'm surprised as 1) I've not had these sorts of errors before, 2) they're not words, and words must be heavily weighted for in the OCR engine (I'd have thought).