JST CREST CoAugmented Sounde Media Communication Project

Our project aims to create basic techniques for augmenting speech production and hearing functions beyond physical constraints based on cooperative actions between a user and a system. We will develop cooperatively augmented speech production and hearing techniques, e.g., low-latency real-time processing, system behavior control using user’s multi-modal behavior signals, and intentional system control through human-machine interaction in a data-driven framework, for making it possible to recover and augment our speech production and hearing functions.