Neural and Statistical Machine Translation: Confronting the State of the Art

This paper presents a comparison of neural and statistical machine translation from the perspective of their emergence, development, and the challenges they currently face. The aim is to provide an overview of the state of the art of two of the most recently used automatic translation systems, confronting their development, as well as the results they demonstrate in real translation tests. The methodology of analysis is based on the translation of a medical corpus using both machine translation models. The results largely reflect the common achievements and issues that persist in these models.