Two-parameter model of synthetic distortions in the problem of assessing the readability of distorted texts

The paper proposes a two-parameter model of random synthetic text distortions. The model provides for distortions both at the level of text symbols and at the level of words. The distortions introduced by the proposed model are close to the distortions that occur when recognition systems (automatic speech recognition and optical character recognition) operate in noise. The model is used to study the redundancy of text in natural languages and to analyze the possibility of its automatic processing under noise conditions. Estimates of the readability of text distorted using the proposed model are obtained. © The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2025.