A new AI-powered translation program is now allowing for the instant translation of tens of thousands of digitalized tablets. In a report by The Times of Israel, archaeologists and computer scientists hope this program can assist in cuneiform interpretation.
Currently, more than half a million clay tablets are inscribed with Cuneiform. The problem is that there are very few Akkadian readers, as the language has been dead for over 2,000 years. But with this Google Translate-type program, there is new hope to remove the language barrier.
Gai Gutherz, a computer scientist who was part of the team that developed the program said, “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.”
Gutherz continued, “I can just use the algorithm to understand and discover what the past has to say.” This project was the thesis project for Gutherz’s master’s degree at Tel Aviv University. Last month, the team went on to publish a research paper, describing its neural machine translation from Akkadian to English.
And this is what gives the team’s program its similarity to Google Translate. As that program also utilizes Neural machine transition. The way this works is that words are turned into a string of numbers. Then, the program uses a complex neural network to generate the sentence in another language.
As for the Akkadian language in question, it was written and spoken in both Mesopotamia and the Middle East from 3,000 BC to 100 AD. Much like Latin was for the Roman Empire, it was the common tongue used to communicate amount the many diverse regions of the area.
Like Sumerian, Akkadian was written using cuneiform and is considered to be among the earliest written languages discovered. As Gai Gutherz puts it, what makes this exciting is a new world has now opened like never before. “Translating all the tablets that remain untranslated could expose us to the first days of history, to the civilization of those people, what they believed in, what they were talking about, what they were documenting.”
Finally, the team is sharing its open-source research online in hopes that experts that work on other dead languages will create their own translation programs and bring to life cultures long gone.