Simultaneous interpreting is more than just real-time translation—it requires processes like summarization, simplification, generalization, and paraphrasing. While these strategies help interpreters manage cognitive load, they also typically enhance audience comprehension. In this talk, I will present findings from a dataset of student interpreting, discussing the challenges of annotation, key characteristics of the data, and notable linguistic patterns. I will also outline our progress toward automatic alignment methods and their potential applications. Beyond dataset analysis, I’ll explore broader questions: Can we improve the automatic evaluation of simultaneous interpreting? How can these insights help train better interpreters and refine machine translation systems?
*** The talk will be delivered in person (MFF UK, Malostranské nám. 25, 4th floor, room S1) and will be streamed via Zoom. For details how to join the Zoom meeting, please write to sevcikova et ufal.mff.cuni.cz ***