Researchers Teach Computers to Read Ancient Arabic and Buddhist Manuscripts

The manuscripts are first digitized and processed to improve text clarity and remove visual defects. That step is especially important for Arabic, where diacritical marks placed above or below letters can change their meaning. The text is then recognized by the software. Because Arabic is written in a connected script, each letter can take one of four different forms depending on its position within a word.

According to Alikberov, researchers today increasingly use terms such as “digital Oriental studies,” “digital history,” and “digital archaeology.” Over time, however, he believes the word “digital” will disappear because these tools will become a natural part of every scientific discipline.