Posted on 05/07/2023 1:14:46 PM PDT by SunkenCiv
Israeli experts have created a program to translate an ancient language that is difficult to decipher, allowing automatic and accurate translation from cuneiform characters into English.
Researchers at Tel Aviv University (TAU) and Ariel University have developed an artificial intelligence model that can automatically translate Akkadian text written in cuneiform into English.
Experts in Assyriology have spent years studying cuneiform, one of the earliest known writing systems, in order to comprehend ancient Mesopotamian texts.
Dr. Shai Gordin of Ariel University and Dr. Gai Gutherz, Dr. Jonathan Berant, and Dr. Omer Levy of TAU trained two versions of the AI model – one that translates Akkadian from representations of cuneiform signs in Latin script and another that translates from unicode representations of the signs.
With a score of 37.47 on the Best Bilingual Evaluation Understudy 4 (BLEU4), the first version—which uses Latin transliteration—produced better results in this study. This means that the model can produce translations that are on par with those produced by an average machine translation from one modern language to another. Given that there is a cultural gap of more than 2,000 years in translating ancient Akkadian, this is a noteworthy accomplishment...
In 2020, the same group of researchers created an AI model called "the Babylonian Engine." The contemporary model is supposedly a better and reworked version of it.
Historians note that hundreds of thousands of clay tablets from ancient Mesopotamia, written in cuneiform, have been found by archaeologists, far more than can be translated by the limited number of experts who can read them.
(Excerpt) Read more at arkeonews.net ...
It’s all Greek to me...
Here’s an instant where AI seems appropriately useful. I wonder if the AI model shows biases in its interpretation.
Now how about Cretan Linear A?
allowing automatic and accurate translation
I once heard a lecture by an Israeli scholar who specializes in the Ancient Near East. He said that none of the experts in the field knows all of the ancient languages (there are too many of them).
If the texts are government decrees or other official documents, maybe the guiding principle should be "close enough for government work."
Fascinating. I am prepared to be surprised in any and all directions.
Maybe AI could be unleashed on the Venona Project intercepts, only a small fraction of which were ever deciphered. If the originals still exist. There are innumerable corners of the history of the period that could be illuminated.
Maybe the English comes out like a Google Translate product which is useful for giving you the gist of it.
I think the machine translation would be used to identify which of the hundreds of thousands of untranslated tablets should be translated by a human.
Amazing. The ancient scribes must’ve been really patient with whoever was dictating this stuff, or Vice Versa. There’s a lot of effort in carving all those symbols, it seems.
I wonder if they got paid by the symbol…
Given that there is a cultural gap of more than 2,000 years in translating ancient Akkadian, this is a noteworthy accomplishment...
—
Akkadian Empire 2334 - 2154 BC Can you say over 4,000 years?
selections from three keywords (mostly Epigraphy and Language), sorted:
Where is the Rosetta Cuneiform Stela?
They should give it a crack at the Voynich manuscript while they’re at it.
In before “be sure to drink your Ovaltine.”
“Now how about Cretan Linear A?”
Yes, I hope that’s next on the AI translation of texts. Would love to have that mystery solved. Dying of curiosity, I am.
This Post #1 is not an image uf cuneiform, is it? Looks more pictorial, like heiroglyphics, eh>?
Microsoft supposedly has an AI that can read any language, even if it never was taught how........................
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.