DSC internal talk in February

Prof. Satoshi Nakamura gave a lecture in February.

The details are as follows.

====================

Prof. Satoshi Nakamura (Augmented Human Communication Laboratory)

TITLE: 

Advanced in Spoken Language Processing Research

Abstract:

The past and current status of research in spoken language processing will be introduced based on the speakers’ footsteps and technological advances.　Speech processing, such as speech recognition and speech synthesis, which were still considered dream technologies in the 1980s, has made remarkable progress with the introduction of probabilistic models and advances in neural networks and has now spread to the point where they are part of people’s everyday tools. Machine translation also had its twists and turns, such as a report in the U.S. that said the government should not invest in research for machine translation, but significant progress has been made with the emergence of word2vec, which allows word representation in a continuous space, and various deep learning models. Today, large-scale language models are evolving much faster than expected, and the best results to date are being rewritten one after another. This presentation will showcase these technological advances by overlaying the speaker’s footsteps.