Technology

A new tool translates 4000-year old stories using machine learning

An ancient egyption inscription. Picture by special arrangement.

An ancient egyption inscription. Picture by special arrangement.  

(Subscribe to our Today's Cache newsletter for a quick snapshot of top 5 tech stories. Click here to subscribe for free.)

Ancient Egyptians used hieroglyphs over four millennia ago to engrave and record their stories. Today, only a select group of people know how to read or interpret those inscriptions.

To read and decipher the ancient hieroglyphic writing, researchers and scholars have been using the Rosetta Stone, an irregularly shaped black granite stone.

In 2017, game developer Ubisoft launched an initiative to use AI and machine learning to understand the written language of the Pharoahs.

The initiative brought researchers from Australia’s Macquarie University and Google’s Art and Culture division togther.

The team worked on five-staged translation process of four millennia old text using machine learning. It involved pre-processing the inscription, sequencing the glyph, classifying it, matching the word to dictionary, and then finally, translating the text.

 

The group trained over 80,000 drawings of hieroglyphs on the machine learning drawing tool called Workbench. The images were trained on another tool called Psycle.

“Ubisoft's Hieroglyphics Initiative is developing a machine learning based tool – the Workbench Tool - that will create a clear workflow to process images of hieroglyphs, identify signs and translate the glyphs, a bit like Google Translate does,” Dr. Alexandra Woods, a senior Egyptology lecturer at Macquarie University, told about working on translation process.

Fabricius lets users learn the language of the Pharoahs. Picture by special arrangment.

Fabricius lets users learn the language of the Pharoahs. Picture by special arrangment.  

 

Macquarie’s Department of Ancient History signed up as Principal Academic Partner for The Hieroglyphics Initiative. And professionals at Google engaged in transitioning Hieroglypics Initiative workbench data on to the search giant’s platform.

This has led to a new tool, built by Google, that allows anyone discover the ancient writing form through three dedicated pathways.

The ancient language decoding AI tool Fabricius takes the curious-minded people through a process of learning the olden writing form through short educational video.

The language tool then invites the person to play and translate their own words and messages into hieroglyphics. The translated texts can be shared with their friends and family.

Fabricius then, slowly progresses the learning quest to newer avenues of academic research.

“So far, experts had to dig manually through books upon books to translate and decipher the ancient language -- a process that has remained virtually unchanged for over a century,” Chance Coughenour, Programme Manager at Google Arts & Culture, said in a blog post.

“Fabricius includes the first digital tool that decodes Egyptian hieroglyphs built on machine learning.”

The tool has also been launched as open source to support further developments in the study of ancient languages.

Google has used its machine learning technology AutoML Vision to create a model that can make sense of the ancient Egyptian language.

The search giant says that what used to take a team of data scientists to do, can now be developed using AutoML Vision by easily training a machine to recognise all kinds of objects.

Fabricius is available in English and Arabic, and is named after the father of epigraphy, the study of ancient inscriptions.

Recommended for you
This article is closed for comments.
Please Email the Editor

Printable version | Aug 15, 2020 7:31:36 PM | https://www.thehindu.com/sci-tech/technology/a-new-tool-translates-4000-year-old-stories-using-machine-learning/article32098001.ece

Next Story