Voice Recognition Technology Potential for a Complex Map Display System(Taligenkaenning i ett Komplext Presentationssystem foer Laegeskartor)

摘要

The present study was designed to compare mouse, keyboard, and voice input modesfor commands to a computer graphic map presentation system. In addition, it compared performance for both pre-composed and user-composed queries. Finally, the study considered the performance of an American speech recognition system, in recognizing commands spoken in English by non-native speakers, as well as commands made using Swedish. Results indicated a modes general advantage for voice as an input mode. They also indicated that performance was a function of the type of query, and that the effects of query type depended, in turn, on the mode of input. While voice showed a strong advantage for pre-composed queries, mouse and voice were comparable for user-composed queries, with the keyboard condition resulting in universally poor performance. Finally, although native speakers of Swedish generally performed well when issuing commands in English, the speech recognition system's performance for Swedish commands was comparatively poor.