‘World’s oldest texts’ decoded by AI as groundbreaking tech solves mysteries of 5,000-year-old tablets worn down by time
ANCIENT texts scribbled on 5,000-year-old tablets can be finally decoded thanks to the groundbreaking AI technology.
The new research used “Google Translate”-like program to uncover the mysteries behind humanity’s earliest forms of writing.
A team from Martin Luther University Halle-Wittenberg, Johannes Gutenberg University Mainz, and Mainz University of Applied Sciences has unveiled an AI system capable of deciphering ancient cuneiform texts.
The groundbreaking research used nearly 2,000 cuneiform tablets, but there are reportedly one million of such tablets in the world.
Most of them originated from Mesopotamia, a historical region in present-day Iraq, known as the cradle of the civilization.
Many of them are over 5,000 years old, making them mankind’s oldest surviving written records.
The ancient scripts cover a wide range of topics from shopping lists to court rulings.
Hubert Mara, one of the study’s authors, said: “The tablets provide a glimpse into mankind’s past several millennia ago.
“However, they are heavily weathered and thus difficult to decipher even for trained eyes.”
The ancient language was inscribed in form of symbols, signs and wedges on wet clay tablets.
Thousands of years have worn down these tablets, making them hard to read.
The traditional technology can scan and translate manuscripts and other two-dimensional materials.
But it doesn’t account for lighting and colour distractions that can impact 3D objects like tablets.
“OCR (Optical Character Recognition) usually works with photographs or scans.
“This is no problem for ink on paper or parchment.
“In the case of cuneiform tablets, however, things are more difficult because the light and the viewing angle greatly influence how well certain characters can be identified,” said co-author Ernst Stötzner.
But the new system eliminates the need for optimal lighting, and even background knowledge of the ancient language when translating the text.
“What’s so amazing about it is that I don’t need to understand Akkadian at all to translate [a tablet] and get what’s behind the cuneiform,” said Gai Gutherz, a computer scientist who was part of the team that developed the program.
“I can just use the algorithm to understand and discover what the past has to say.”
The prototype of the software can reliably discern symbols from two languages.
One of them is Akkadian that hasn’t been spoken or written for over 2,000 years.
There are a total of twelve cuneiform languages known to exist.
“Translating all the tablets that remain untranslated could expose us to the first days of history, to the civilisation of those people, what they believed in, what they were talking about, what they were documenting,” said Gutherz.
Future developments could see the extension of the technology’s use on other 3D scripts such as weathered inscriptions found on headstones in cemeteries.
