Hidenori Itoh

Bayesian-Based Inference of Dialogist's Emotion for Sensitivity Robots

We describe a method for sensitivity communication robots which infer their dialogists emotion. The method is based on the Bayesian approach: by using a Bayesian modeling for prosodic features. In this research, we focus the elements of emotion included in dialogists voice. Thus, as training datasets for learning Bayesian networks, we extract prosodic feature quantities from emotionally expressive voice data. Our method learns the dependence and its strength between dialogists utterance and his emotion, by building Bayesian networks. Bayesian information criterion, one of the information theoretical model selection method, is used in the building Bayesian networks. The paper finally proposes a reasoner to infer dialogists emotion by using a Bayesian network for prosodic features of the dialogists voice. The paper also reports some empirical reasoning performance.

A bayesian approach to emotion detection in dialogist’s voice for human robot interaction

This paper proposes a method for sensitivity communication robots which infer their dialogist’s emotion. The method is based on the Bayesian approach: by using a Bayesian modeling for prosodic features. In this research, we focus the elements of emotion included in dialogist’s voice. Thus, as training datasets for learning Bayesian networks, we extract prosodic feature quantities from emotionally expressive voice data. Our method learns the dependence and its strength between dialogist’s utterance and his emotion, by building Bayesian networks. Bayesian information criterion, one of the information theoretical model selection method, is used in the building Bayesian networks. The paper finally proposes a reasoner to infer dialogist’s emotion by using a Bayesian network for prosodic features of the dialogist’s voice. The paper also reports some empirical reasoning performance.

Generating Interactive Facial Expression of Communication Robots Using Simple Recurrent Network

To improve face-to-face interaction with robots, we developed a model for generating interactive facial expressions by using a simple recurrent network (SRN). Conventional models for robot facial expression use predefined expressions, so only a limited number of expressions can be presented. This means that the expression may not match the interaction and that the person may find the expressions monotonous. Both problems can be overcome by generating expressions dynamically. We tested this model by incorporating it into a robot and comparing the expressions generated with those of a conventional model. The results demonstrated that using our model increases the diversity of face-to-face interaction with robots.

A Biphase-Bayesian-Based Method of Emotion Detection from Talking Voice

This paper propose a Bayesian-based method of emotion detection from talking voice. Development of a entertainment robot and joyful communication between human and robot have given us the motivation for a computational method for robot to detect its dialogists emotion from his talking voice. The method is based on the Bayesian networks which represent the dependence and its strength between dialogists utterance and his emotion, by using a Bayesian modeling for prosodic feature quantities extracted from emotionally expressive voice data. In this paper, we propose a biphase inference method using the Bayesian networks. This inference method has two steps: to reduce the choice of emotion at the first step and to infer a certain emotion reliably from little choice at the second step. The paper also reports some empirical reasoning performance of this method.

Evolution of Migration Behavior with Multi-agent Simulation

We describe an artificial ecosystem consisting of five areas, evolving artificial creatures (called agents), and foods for the agents. The ecosystem was constructed for an analysis of migration behavior of the Monarch butterfly. We also report our simulation results on the emergence of the migration biology pertaining to the Monarch butterfly. We use real temperature data as an environmental parameter to define the environment in the field. To adapt to the environment, the agents have temperature sensors. We conducted two experiments using this ecosystem. The results show that the biology of the Monarch butterfly has been well modeled by the ecosystem and our evolutionary method.

A neural-based approach to facial expression mapping between human and robot

This paper proposes a neural-based method to map facial expressions between human and robot. We applied the method to a sensitivity communication robot, Ifbot, which has been developed by our industry-university joint research project. The method enables the robot to imitate an arbitrary human facial expression. The paper describes the feature extraction from face image, and proposes neural network based parameter matching between human facial expression and Ifbots facial expression mechanism. This paper also reports the evaluation of the facial expression transmission performance of Ifbot with the proposed system. The evaluation shows the effectiveness of emphasizing emotional expressions and the possibility of using Ifbot as an agent for distance communication.