DeepL Traduttore e didattica della traduzione dall’italiano in tedesco - Alcune valutazioni preliminari

What effects can the availability of a tool such as DeepL Translator have on the teaching of translation from Italian into German? This paper examines the most relevant factors to be taken into account, beyond simple standardised measurements. Starting from samples contained in two theses of Bachelor and Master's degree courses in translation, we explore quality levels of DeepL Translator’s raw output (with some reflections on BLEU scores) and successive improvements achievable with a simple kind of post-editing (without formal check-lists). Our findings confirm that this application provides quite promising results even without domain-related adaptation. Although some residual failures to be addressed in post-editing are insidious and challenging, DeepL Translator is likely to become an invaluable tool for more advanced and efficient practices in teaching and training. Accordingly, post-editing should become an essential training activity.