Carletta, J.; Caley, R.; and Isard, S. 1993. A system architecture for simulating time-constrained language production. Technical Report rp-43, Human Computer Research Centre, University of Edinburgh. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Computer-generated Dialogue as a Resource-bounded Activity - Donaldson, Cohen (Correct)
....example of such risk is evidenced by self repairs, where a speaker catches himself in mid utterance, e.g. Go to the left uh the right of the swamp. This research has motivated the development of a systemic natural language generation system that is capable of anytime language processing (Carletta et al. 1993). The system is seeded with default values for all possible output, and so if the system is interrupted prematurely, it can always say something by using its defaults, or by purposely inserting a hesitation if a certain level of solution quality has not been achieved. Fillers can easily be taken ....
....example, if after the systems asks G 3 the student happens to answer his own original question and responds Oh, I see, I can t take CS100 because I ve already taken CS102, then the advisor can drop all of its goals with respect to answering the student s question, ending the dialog. Discussion Carletta et al. 1993) suggest another possibility of for taking advantage of a performance profile: along with a lower quality bound of hesitate , an upper quality bound of speak can be used to let the system decide when to speak in cases where it can speak, but is not absolutely required to do so. Such cases ....
Carletta, J.; Caley, R.; and Isard, S. 1993. A system architecture for simulating time-constrained language production. Technical Report rp-43, Human Computer Research Centre, University of Edinburgh.
Turn-taking in Discourse and its Application to the Design.. - Donaldson, Cohen (1996) (Correct)
....taking: it is worth studying how information gathering subdialogues can be treated as interruptions and how the use of time bounded persistent goals can provide for a richer set of possible dialogue structures. Relevant computational work on resource bounded language generation has been done by Carletta et al. 1993) and Walker (1993) Carletta et al. are partly interested in the cognitive plausibility of language production, and treat it as a high risk activity, where it is acceptable for a speaker to make a mistake because it is often possible to quickly repair this mistake. Their notion of anytime ....
Carletta, J.; Caley, R.; and Isard, S. 1993. A system architecture for simulating time-constrained language production. Technical Report rp-43, Human Computer Research Centre, University of Edinburgh.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC