Skip to content

Technology

Technology powering spotless OCR

Photographed text or handwritten notes: UglyWriting gives you a clean document ready for review.

Accuracy benchmarks by use case

  • 99.7 % School notes (scanned)
  • 98.5 % School notes (smartphone photo)
  • 99.3 % Old book photograph
  • 99.9 % Screenshot

Demo

See for yourself

Four real documents, sent to UglyWriting as they were. The transcriptions below have not been touched up: the remaining errors are visible.

School notes (scanned)

Reads the handwriting despite the notebook lines, rejoins words split across lines and ignores the drawing around the last paragraph.

99.7% accuracy 2 errors out of 591 characters.

Original image
Original document — School notes (scanned)
UglyWriting transcription

Ugly Writing

Wikipedia says that OCR means Optical Character Recognition or Reader and is defined as the electronic or mechanical conversion of images of text into machine-encoded text.

Ugly Writing aims to be your partner in any task that includes OCR. Are you a student looking for a efficient way to store your hand-written notes? Do you work in a library and need to sort old books? Are you writing a novel? Have you found your grandma's diary?

As Ugly Writing is hosted in France, you are covered by the very strict GDPR. Your data are yours and will stay yours. We will not sell 'em.

School notes (smartphone photo)

Even from a quick snapshot with a pen lying on the page, the text is faithfully restored, notebook line breaks included.

98.5% accuracy 9 errors out of 591 characters.

Original image
Original document — School notes (smartphone photo)
UglyWriting transcription

Ugly Writing

Wikipedia says that OCR means
Optical Character Recognition or
Reader and is defined as the electronic
or mechanical conversion of images of text
into machine-encoded text.

Ugly Writing aims to be your partner
in any task that includes OCR. Are
you a student looking for a
efficient way to store your hand-
written notes? Do you work in a
library and need to sort old books?
Are you writing a novel? Have
you found your grandma's diary?

A's Ugly Writing is hosted
in France, you are covered
by the very strict GDPR.
Your data are yours and
will stay yours. We will not sell

Old book photograph

Preserves layout and footnotes despite tiny writing, aged paper, a near-century-old print, and a smartphone photo.

99.3% accuracy 11 errors out of 1626 characters.

Original image
Original document — Old book photograph
UglyWriting transcription

LIVRE PREMIER

I. — LA CIGALE ET LA FOURMI¹

La Cigale, ayant chanté
Tout l'été,
Se trouva fort dépourvue
Quand la bise² fut venue :
Pas un seul petit morceau
De mouche ou de vermisséau.
Elle alla crier famine
Chez la Fourmi sa voisine,
La priant de lui prêter
Quelque grain pour subsister
Jusqu'à la saison nouvelle.
« Je vous paierai, lui dit-elle,
Avant l'ouït³, foi d'animal,
Intérêt et principal⁴. »
La Fourmi n'est pas prêteuse :
C'est là son moindre défaut⁵.

1. Esope, 154. Le même sujet a été traité par Babrius. On en rencontre de nombreuses imitations chez les Latins et dans les poètes français qui sont antérieurs à La Fontaine.

2. Bise, signifie proprement le vent du nord. Il est employé ici pour le temps froid, l'hiver.

3. C'est le mois d'août (Auguste), époque de la moisson. La Fontaine dit ailleurs :

Remuez votre champ dès qu'on aura
(fait l'ouït.
(Liv. V, fab. ix.)

4. L'engagement est complet : le serment, foi d'animal, la date du remboursement, avant l'ouït, la somme intégrale, intérêt et principal, rien n'y

manque. Les emprunteurs ne sont pas avares de promesses ; mais, selon Franklin, ceux qui font des dettes ont toujours le mensonge en croque.

5. Ce vers est obscur. La Fontaine ne veut pas dire que le plus petit défaut de la fourmi soit de ne pas prêter, mais que le prêt est ce qu'il y a de plus contraire à ses habitudes, la vertu qu'elle pratique le moins, et ironiquement son moindre défaut. — Ce sens donné à moindre est incontestable dans ce vers de Molière (École des Maris, acte I, scène vi) :

Je coquette fort peu, c'est mon moindre
(talent.

C'est-à-dire le talent one je possède le moins.

Screenshot

Text, bold, the video caption and reference marks are restored; the page header and the video thumbnail are kept as images.

99.9% accuracy 2 errors out of 1569 characters.

Original image
Original document — Screenshot
UglyWriting transcription

Wikipedia page header, kept as an image

Form Wikipedia, the free encyclopedia

Video thumbnail: a portable scanner reading a printed page
Video of the process of scanning and real-time optical character recognition (OCR) with a portable scanner

Optical character recognition or optical character reader (OCR) is the electronic or mechanical conversion of images of typed, handwritten or printed text into machine-encoded text, whether from a scanned document, a photo of a document, a scene photo (for example the text on signs and billboards in a landscape photo) or from subtitle text superimposed on an image (for example: from a television broadcast).[1]

Widely used as a form of data entry from printed paper data records – whether passport documents, invoices, bank statements, computerized receipts, business cards, mail, printed data, or any suitable documentation – it is a common method of digitizing printed texts so that they can be electronically edited, searched, stored more compactly, displayed online, and used in machine processes such as cognitive computing, machine translation, (extracted) text-to-speech, key data and text mining. OCR is a field of research in pattern recognition, artificial intelligence and computer vision.

Early versions needed to be trained with images of each character, and worked on one font at a time. Advanced systems capable of producing a high degree of accuracy for most fonts are now common, and with support for a variety of image file format inputs.[2] Some systems are capable of reproducing formatted output that closely approximates the original page including images, columns, and other non-textual components.

Accuracy: share of characters correctly restored, measured by edit distance between the transcription and the reference text (spaces and formatting markup excluded). These examples show how effective UglyWriting is. Results vary according to image quality and handwriting legibility.

Designed for every situation

Illustrations preserved

Diagrams, photos, and annotations keep their position so the document stays meaningful.

Rebuilt tables

Cells, headers, and merges are recreated in semantic HTML or DOCX table objects.

Easy-to-edit exports

Download DOCX or responsive HTML for quick review and edits.

A better output than general-purpose OCR APIs

Our benchmarks show how UglyWriting stands out against generic solutions.

Figures based on our 2025 internal multilingual test suites across a panel of competitors.

Evaluated metric UglyWriting Competitors
Document fidelity ★★★★★ (near perfect) ★★★☆ (good)
Handwriting extraction Advanced model, dedicated training Experimental feature
Tables & illustrations Native tables + embedded images Image-to-text conversion only

Experience the difference on your documents

Scan or photograph a document: our pipeline delivers reliable text ready for refinement.