Associate Professor Yoshihiko Nankaku
Based on information processing technology for voice and images, we are promoting a wide range of research on multimedia signal processing with the aim of "realization of intelligent machines that interact like humans". Based on statistical machine learning and deep learning, we explore new theories related to speech recognition, speech synthesis, spoken dialogue systems, and image recognition, and investigate the mechanisms behind them to realize high performance and system integration. Enables application to the environment. Specifically, technologies that can be used in familiar environments such as voice dialogues with smartphones and smart speakers, voice guidance for digital signage and car navigation, security systems and entertainment using audiovisual information, and communication support for people with disabilities is. In the future, we are conducting research every day with the aim of realizing machines that can recognize more accurately than humans, machines with richer expressions than humans, and technologies that expand human communication skills.
Link to Laboratory Website
Copyright© NITech Frontier Research Institutes, Inc. All Rights Reserved.