archeology
Researchers call it a major step in the preservation of Mesopotamian cultural heritage.
Scientists from Israel’s Tel Aviv University and Ariel University say they have used artificial intelligence to translate fragments of ancient cuneiform script written on stone tablets into English with high accuracy. They call the project “another major step towards the preservation and dissemination of ancient Mesopotamian cultural heritage.”
Academics published the first neural machine translation from Akkadian to English in the May issue of the magazine. PNAS Nexus. The results, they say, are “comparable to those produced by an average machine translation from one modern language to another.” Archeo News.
In the past 200 years, archaeologists have uncovered hundreds of thousands of documents telling the history of ancient Mesopotamia, most of them written in Sumerian or Akkadian, the authors explained. However, due to its sheer volume, the small number of experts who can read it, and the fact that most documents are fragmentary, most of it remains untranslated. In addition, cuneiform is polyvalent, with many different types of characters, and even the names of people and places can be written as complex sentences.
“First and foremost, we do not believe that AI will replace the work of philology,” says one of the authors of the Digital Past Laboratory, Ariel University School of Land Studies and Archeology, Israel. Luis Saenz said in an email: artnet news. “We want to speed up the process. Our hope is that eventually AI will be able to read cuneiform in the future for both Assyrian and non-Assyrian scholars.”
This is just the latest example of scientists using modern tools to work with the oldest materials.Researcher at the University of Kentucky Developed an AI system To read scrolls burned during the eruption of Mount Vesuvius in AD 79, Italian archaeologists working on a robot Use AI to reconstruct ancient artifacts from scattered debris.
“Of course, this model has limitations,” Saenz says. “Ancient languages are difficult to translate because they are only fragments of text and therefore lack context. To continue and improve results, we need more tools to digitize published data, and a user-friendly web-based platform for the general public is also important.”
Other trending articles:
Smithsonian’s National African Art Museum director resigns after less than two years, citing ‘resistance and backlash’
‘We’re not all minimalists who love IKEA’: Historian and author Michael Diaz Griffiths on the resurgence of young antique collectors
The late billionaire Heidi Horten’s controversial jewelery first auction was a huge success, raising $156 million
Airbnb hosts got more than they bargained for with guests’ quirky art exchanges: The mystery is trending on TikTok
‘Her Story’ of Art, Not Patriarchal Art History: Judy Chicago on Why She Dedicates a New Show to the 80 Female Artists Who Inspire Her
An artist asked ChatGPT how to make the popular Memecoin.The result is ‘TurboToad’ and people are betting millions on it
Elderly man spray-paints Miriam Khan painting at Paris museum after right-wing censorship failure
The Netflix series Transatlantic dramatizes efforts to evacuate artists from France during World War II.Here’s what actually happened in real life
Follow Artnet News on Facebook:
Want to stay ahead of the art world? Subscribe to our newsletter for the latest conversation-forward news, eye-popping interviews, and incisive critical views.
