AI Already Did It › Productivity & Office › PDF Tools

Glyph OCR

Name: Glyph OCR
Author: Administrator

Short description: Offline desktop OCR for Linux Mint that turns images, screenshots, scanned PDFs and about 140 other document types into editable text, powered by xberg (formerly kreuzberg).

License	MIT
Generated by	Anthropic Claude Fable 5.1 —
Platforms	linux
Rating	0/100 (0 ratings) · worked for 0, didn't for 0
Downloads	0
Latest version	1.0.0 (2026-09-09)
Uploader	Administrator
Estimated generation cost	518,000 tokens (≈ $7.77) — saved so far: 0 tokens

Screenshots

Glyph OCR screenshot

Description

Glyph OCR is a desktop text-recognition application for Linux Mint built on the xberg document-intelligence engine. Open a picture, a scanned PDF, an Office document, an e-book, an e-mail or an archive and the recognised text appears in an editable, searchable editor next to a zoomable preview of the page. Word boxes are drawn over the preview, coloured by confidence.

Screen capture OCR: global hotkey, drag a rectangle over any window, text lands on the clipboard. Dark UI captures are inverted automatically.
Clipboard OCR and drag-and-drop of files or images.
Region recognition: select part of a page and recognise only that area, or append several regions.
Recognition modes: auto layout, single block, single column, scattered text, single line or word, vertical text, orientation detection.
Engines: Tesseract (default, 100+ languages, auto-downloaded data), PaddleOCR, TrOCR for handwriting, and opt-in vision-language models (OpenAI, Anthropic, local Ollama).
Preprocessing: DPI, deskew, auto-rotate, denoise, contrast, invert, binarisation.
Documents: multi-page PDF viewer, per-page confidence, tables, metadata, PDF passwords, embedded images, recursive archives.
Output: plain text, Markdown, HTML, Djot; export to TXT, MD, HTML, DOCX, PDF, JSON; batch conversion of folders.
Desktop integration: tray, notifications, menu entry, CLI flags, and a .deb with a self-contained runtime under /opt/glyph-ocr.

Downloads

glyph-ocr_0.1.0_amd64.deb — Linux, 78.2 MB
ocr.tar.gz — Source, 217.9 KB

Original prompt(s)

Prompt 1

We need to create Optical character recognition (OCR) application for my Linux Mint
 - To do that: We need to use https://github.com/xberg-io/xberg, https://docs.kreuzberg.dev/guides/ocr/ (previously knwon as "kreuzberg") library
 - Support every file type xberg/kreuzberg supports
 - Look current OCR softwares: 
    * https://www.screensnap.pro/blog/ocr-software-windows
    * https://vovsoft.com/software/ocr-reader/
    * https://apps.microsoft.com/detail/9n2lpktlr1j3?hl=en-US&gl=US

Research those application before implementing any code

It should be desktop application

Tags: desktop-app, kreuzberg, ocr, paddleocr, pyside6, python, screenshot-ocr, tesseract, text-recognition, xberg

Open in AI Already Did It