Motion Editing Tool for Reproducing Grammatical Elements of Japanese Sign Language Avatar Animation

Tsubasa Uchida, Naoki Nakatani, Taro Miyazaki, Hiroyuki Kaneko, Masanori Sano

Abstract: For deaf and hard of hearing people whose native language is sign language, it is necessary to provide information not only with subtitles but also with sign language. One means of providing information in sign language is to use animated avatars. Therefore, we have developed a system that generates a Japanese Sign Language (JSL) avatar animation from Japanese sentences utilizing Japanese-to-JSL translation and a motion-data-driven animation generation method. In this paper, we propose a motion editing tool that can adjust grammatical elements of JSL by editing each motion data forming a JSL sentence. An evaluation experiment shows that editing the motion speed and blend span of multiple words corresponding to the delimitation of phrases and clauses can reduce the error rate of understanding JSL avatar animations.

External IDs:dblp:conf/icassp/UchidaNMKS23