File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/metho/03/p03-2015_metho.xml
Size: 1,392 bytes
Last Modified: 2025-10-06 14:08:22
<?xml version="1.0" standalone="yes"?> <Paper uid="P03-2015"> <Title>Noriyoshi</Title> <Section position="4" start_page="0" end_page="0" type="metho"> <SectionTitle> 4 Conclusion </SectionTitle> <Paragraph position="0"> We have built a spoken dialogue system based on the results of a WOZ experiment with the aim of achieving a television operation interface easy enough for anybody to use.</Paragraph> <Paragraph position="1"> In the preliminary system operation test, 5 subjects were asked to give some examples of TV programs that they watch at home, and to use this system to see whether they could obtain information in relation to those programs. Results of this test showed that all subjects could access information on desired programs. In a subsequent questionnaire, moreover, all subjects stated that &quot;program selection was easy, and particularly there was no need to know about hierarchical structure of program information.&quot; On the other hand, the test also revealed that some issues remain to be addressed in speech recognition but that a favorable evaluation could be obtained from all subjects with regard to television operations via spoken dialogue. We are currently conducting even more detailed experiments to demonstrate the usefulness of a spoken dialogue interface for television control and to examine problem areas.</Paragraph> </Section> class="xml-element"></Paper>