Coffee Maker Robot Based on Simple Vocabulary and Partially Observable Markov Decision Process (POMDP)

Widodo Budiharto, Chiharu Ishii
2017 ICIC Express Letters  
Many recent developments of assistive robotics with face and speech recognition systems have been seen in order to interact with users naturally. In this research, we would like to propose Intelligent Coffee Maker Robot with the speech recognition based on Indonesian language using small vocabulary and statistical dialogue systems. This kind of robot can be used in the office, supermarket or restaurant. In our scenario, robots will recognize the user's face and then accept commands from the
more » ... to do an action, specifically in making 3 types of coffee, espresso, double espresso and latte, where the answer from the user will be processed to the Google Translator and the result will be compared with simple vocabulary in Indonesian language. The main problem here is to know the intention of users about how sweetness of the coffee. The intelligent coffee maker robot should conclude the user's intention through conversation under unreliable automatic speech in noisy environment. In this paper, this spoken dialog problem is treated as a partially observable Markov decision process (POMDP). We describe how this formulation establishes a promising framework by empirical results. The dialog simulations are presented which demonstrate significant quantitative outcomes.
doi:10.24507/icicel.11.04.757 fatcat:44nc6mhcvrffjjsjd34hd4uicu