Scenario speech assignment technique for instant casting movie system
Shinichi Kawamoto, Yoshihiro Adachi, Yamato Ohtani, Tatsuo Yotsukura, Shigeo Morishima, Satoshi Nakamura · Institutional Repositories DataBase (IRDB) · 2009
In this paper, we propose an improved Future Cast System (FCS) that enables anyone to be a movie star while retraining their indivisuality in terms of how they look and how they sound. The proposed system produces voices that are significantly matched to their targets by integrating the results of multiple methods: similar speaker selection and voice morphing. After assigning one CG character to the audience, the system producesvoices in synchronization with the CG character's movement. We constructed the speech sunchronization system using a voice actor database wiht 60 different kinds of voices. Our system achieved higher voice similarity than conventional system; the preference score of our system was 56.5% over other conventional system.