This project builds on the premise that most readability scores are outdated as they do not work on modern text genres and are often language-specific. Eye-tracking data provides information about the linguistic and cognitive processes occurring during text comprehension and can therefore be used as a proxy to determine the readability level of a text. Moreover, we now have models that can accurately predict eye-tracking features for reading, making the need for real-time data obsolete. The goal of this project is to develop a multilingual text readability score based on synthesized eye-tracking data. This score will then be compared to traditional scores and can be evaluated against datasets of standardized language assessment tests or text simplification.