Abstract | ||
---|---|---|
This paper discusses how to build a system that can engage in a mixed-initiative human-machine spoken dia- logue in which system utterances sometimes overlap with user utterances and vice versa. In the method, a mod- ule that incrementally understands user utterances and another module that incrementally generates system ut- terances work in parallel, and the timing of taking and releasing the dialogue initiative is decided according to the understanding of user utterances and the content of the system utterances. This method enables the system to respond when the user holds the dialogue initiative and is speaking, and enables the system to react to the user's barge-ins when it holds the initiative and is speaking. An experimental system called DUG-1 is also presented. This paper proposes a method for handling rich turn- taking in a mixed-initiative spoken dialogue system. It also presents an experimental system called DUG-1. DUG-1 achieves rich turn-taking through the incremental under- standing of user utterances and the incremental generation of system utterances. Rich turn-taking is also achieved by having the understanding module and the generation module work in parallel. We first discuss the necessity of handling rich turn- taking, and then present a method for handling it. Finally, we present an experimental system DUG-1. |
Year | Venue | Field |
---|---|---|
1999 | EUROSPEECH | Turn-taking,Computer science,Artificial intelligence,Natural language processing,Versa |
DocType | Citations | PageRank |
Conference | 4 | 0.60 |
References | Authors | |
12 | 8 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mikio Nakano | 1 | 488 | 61.92 |
Kohji Dohsaka | 2 | 173 | 18.38 |
Noboru Miyazaki | 3 | 68 | 8.12 |
Jun-ichi Hirasawa | 4 | 53 | 5.95 |
Masafumi Tamoto | 5 | 28 | 4.41 |
Masahito Kawamori | 6 | 7 | 3.99 |
Akira Sugiyama | 7 | 12 | 1.41 |
Takeshi Kawabata | 8 | 296 | 51.73 |