Artificial Intelligence Can Now Read and Decipher Ancient Texts

Researchers at the University of Notre Dame are using AI to analyze ancient manuscripts.

Experts say there is a vast amount of material that has survived to this day but has never been read. 

As such, researchers at the University of Notre Dame are developing an artificial neural network to read complex ancient handwriting to improve the capabilities of deep learning transcription.
Walter Scheirer, an Assistant Professor in the Department of Computer Science and Engineering at Dennis O. Doughty, said, “What we’ve set out to do is automate transcription in a way that mimics the perception of the page through the eyes of the expert reader.”

The technology combines traditional machine learning methods with visual psychophysics – a method of measuring the connections between physical stimuli and meta phenomena. For instance, the amount of time it takes for an experienced reader to recognize a specific character, gauge the quality of the handwriting, or identify the use of certain abbreviations.

Scheirer’s team studied digitized Latin manuscripts that were written in the Monastery of Saint Gall in the ninth century. Readers entered their manual transcriptions into a specially designed software interface. The team then measured reaction times during transcription for an understanding of which words, characters, and passages were easy or difficult.

He explained that this data was necessary to decrease the number of errors and create a network more consistent with how a person would perceive text, providing a more accurate, more realistic reading of the manuscript.

Now, the team aims to improve the accuracy of transcriptions, especially when working with damaged or incomplete documents, as well as how to account for illustrations or other accompanying symbols that could lead to confusion.

Thanks to Scheirer’s research, hopefully, we will be able to transcribe and understand millions of historical documents that have been stored away safely in monasteries and libraries all around the world.