Example-based spoken dialog processing for guidance robots

Cheongjae Lee, Seokhwan Kim, Minwoo Jeong, Sangkeun Jung, Donghyeon Lee, Gary Geunbae Lee

Abstract

In this paper, we introduce a natural language interface for human-robot interaction (HRI) to develop building guidance robots. We have developed POSTECH Spoken Dialog System for Guidance Robots (POSSDS-GR). POSSDS-GR consists of automatic speech recognizer, spoken language understanding, dialog manager, natural language generator, and text-to-speech synthesizer. Each module is designed and implemented for making an effective and practical spoken dialog system. Our system is implemented with some dialog examples which are labeled to learn each module. We believe that our spoken dialog system will be a good interface to communicate between human and robot with natural language.