Researchers have developed a groundbreaking artificial intelligence (AI) software capable of deciphering ancient cuneiform tablets. The system eliminates the need for optimal lighting and extensive background knowledge when translating the complex symbols found on these tablets. Instead of relying on photos, the software utilizes 3D models of the tablets, leading to significantly more accurate results compared to previous methods. This breakthrough enables researchers to search through the contents of multiple tablets and compare them with each other.
Cuneiform tablets, estimated to number around 1 million worldwide, are ancient texts that date back over 5,000 years, making them some of humanity’s oldest surviving written records. However, deciphering the intricate symbols has always required optimal lighting conditions and substantial background knowledge.
Hubert Mara, an assistant professor at Martin Luther University Halle-Wittenberg, explained, Up until now, it has been difficult to access the content of many cuneiform tablets at once. You sort of need to know exactly what you are looking for and where.
The new AI system solves this challenge by converting images of the writing and text into machine-readable text. Once transformed into computerized text, the writing becomes more accessible and easier to read and search through. To train the software, the team utilized 3D scans of nearly 2,000 cuneiform tablets, along with additional data, enabling the software to reliably recognize the symbols on the tablets.
Currently, the software’s prototype can accurately discern symbols from two languages. However, the researchers have ambitious plans to expand its capabilities to cover all 12 cuneiform languages known to exist. They also believe the software could assist in deciphering weathered inscriptions, such as those found in cemeteries, which are three-dimensional like cuneiform script.
This breakthrough in AI-assisted cuneiform deciphering holds immense potential for researchers studying ancient languages and civilizations. By eliminating the need for optimal lighting conditions and extensive background knowledge, the software enables a broader exploration of cuneiform tablets’ contents. As a result, researchers can compare multiple tablets and gain deeper insights into ancient cultures and societies.
The AI system’s advancements could not only unlock new knowledge about our past but also pave the way for deciphering other ancient texts. The possibilities for understanding the hidden messages of our ancestors are now within reach, thanks to the marriage of artificial intelligence and ancient script.
