TTAE

Semi-automatic transcription & transliteration of ancient Egyptian

This project continues the OCR-PT-CT project towards implementing a digital toolset to link OCR with transliteration through transcription in a semi-automatic manner. This will be done by matching OCR-ed images from different sources, photographic and facsimile, with the corresponding transliteration strings from the MORTEXVAR database.

A proof of concept

Starting in March 2024, the TTAE project aims to design a transcription-transliteration OCR system of ancient Egyptian hieroglyphic texts through deep learning techniques. The project investigates semi-automatic transcription and transliteration of the hieroglyphic signs as images from different textual sources, mainly facsimiles made by Egyptologists, but also from photographs of the original texts. Stemming off the transcriptions achieved through the OCR-PT-CT project in 2022, the TTAE project will try natural language processing (NLP) techniques for semi-automatic transliteration of the texts. The project is based on the processing and merging of different data sources that form together a corpus of labelled data of a volume which allows training deep learning systems to meet the planned objective. The TTAE project is run by a multidisciplinary team of researchers specialising in artificial vision, signal processing and egyptologists from the Universities of Alcalá, Jaén, Jerusalem, and Rey Juan Carlos University.