Privacy-Secure HRI: Framework of Human-Robot Interaction Protecting User's Privacy

The use of robots that interact with humans in social services has been attracting attention, and the rapid development of AI for automatic dialogue generation has raised expectations for its realization. However, the issue of privacy protection of dialogues between robot and human, which is important when using actual services, has not been taken into account. In this study, we propose a privacy-secure dialogue framework for human-robot interaction (HRI) that comprehensively protects the privacy of dialogue content when a human and a robot engage in dialogue, and we consider privacy protection in all stages from continuous speaker authentication to dialogue generation and dialogue approach. This framework will greatly contribute to the social diffusion of interactive robot services, which are expected to play an active role in public spaces in the future.