Authors:
Takashi Yoshino
;
Ryuichi Nishimura
and
Taku Fukushima
Affiliation:
Wakayama University, Japan
Keyword(s):
Voice collection, Intercultural communication, Parallel texts, Medical field, Web-based system.
Related
Ontology
Subjects/Areas/Topics:
Databases and Datawarehouses
;
Internet Technology
;
Multimedia and User Interfaces
;
Web Information Systems and Technologies
;
Web Interfaces and Applications
;
Web Services and Web Engineering
Abstract:
We have developed a web-based multilingual utterance collection system, named OTOCKER, for the medical field. The purpose of OTOCKER is to act as a voice data collection platform for intercultural communication. Although speech synthesis systems have improved significantly, fluent and smooth speech synthesis is still a problem. Currently, it is difficult to synthesize speech in different languages. In the medical field, in particular, it is important that the intended meaning of spoken words be conveyed effectively along with the different nuances. Therefore, we use an utterance collection system that collects people voices directly. The limitations of this system are (1) an insufficient number of correct sentences pertaining to the medical field and (2) that easy participation in both voice record and collection are difficult. We can solve the first problem by using a system that collects parallel medical texts. The second problem can be solved by using w3voice - a web-based voice-
recording system. This system can run only on a web browser. This paper presents the design of OTOCKER, its prototype, and the results of its trial.
(More)