Haru Ando

Agent-typed multimodal interface using speech, pointing gestures and CG

This paper proposes a sophisticated agent-typed user interface using speech, pointing gestures and CG technologies. An “Agent-typed Interior Design System” has been implemented as a prototype for evaluating the proposed agent-typed interface, which has speech and pointing gestures as input modalities, and in which the agent is realized by 3 dimensional CG (3-D CG) and speech guidance. In this paper, the details of system implementation and evaluation results, which clarified the effectiveness of the agent-typed interface, are described.

Content-generation system using a learner's answering process

A content-generation system that offers adaptive contents for each learner has been developed. In particular, it consists of an assessment function that utilizes a learners process of answering a question. The function can extract event data generated by each learner during an answering process to a question of a test given by PC via a digital pen, and it analyzes the event data in order to extract a learners answering process and pattern. Using this system, each learner can look back at their answering process in addition to look over the final answer. As a result, study efficiency is expected to be improved. Furthermore, the teacher or tutor can create custom-made content for each learner according to the answering process of each learner.

Content-Clipping System Using Multi-modal Analysis

We report on the development of a content authoring system for clipping parts that a user requires from audio-visual e-learning contents. The system can analyze lengthy contents for e-learning by using the contents and information related to the contents. It lets the user easily find parts that he/she wants via the user interface. This system enables users to peruse only the required part of the lesson content, which should improve the efficiency of studying.