andylitalo/grabar-ocr
Manages pipeline to digitize "Grabar" (Classical Armenian) texts
An end-to-end pipeline for digitizing Classical Armenian (Grabar) texts from scanned PDFs into a searchable, translated PostgreSQL database. The system uses PyMuPDF and YOLOv8 for layout detection, fine-tuned TrOCR for Armenian OCR, Claude/GPT-4 for translation, and Apache Airflow for orchestration on a k3s Kubernetes cluster with GPU support.
An Armenian OCR pipeline combining CRAFT for text detection and a deep learning model for text recognition, specifica...
A Python toolkit for performing OCR on Armenian historical documents. It uses pytesseract with multiple preprocessing...
A specialized OCR tool for digitizing historical Armenian newspapers from the Haratch archive. It automates downloadi...
An open-source OCR model trained for Armenian text recognition, supporting Classical, Western, and Eastern Armenian. ...
A Java library that uses Google Cloud Vision API to extract Armenian text from images and saves the output to a .docx...
A web application for Armenian OCR (Optical Character Recognition) built with Elixir and Phoenix. It extracts Armenia...