A Glimpse at the Future of NLP from Spoken Audio Translation – keynote Marco Turchi

Recent advances in deep learning are giving the possibility to address traditional NLP tasks in a new and completely different manner. One of these tasks is spoken language translation, addressed for years by cascading automatic speech recognition and a machine translation system. New trends rely on using a single neural network to directly translate the input audio signal in one language into a text in a different language without intermediate transcription steps.
In this conference, researcher Marco Turchi (Head of the Human Language Technology Machine Translation group at Fondazione Bruno Kessler) will present his recent advancements in end-to-end speech translation and will analyze the benefit of directly accessing the audio in two different tasks: subtitle generation and mitigating the gender bias in spoken language translation. Then, we will have a look at some of the most advanced projects in NLP and will discuss its future with top experts.