Dialog supporting apparatus
View Patent ↗A dialog supporting apparatus which can support an on-going dialog so that the dialog is smoothly completed. The dialog supporting apparatus includes an utterance receiving unit which receives an utterance of a dialog participant and outputs utterance information, an utterance processing unit which translates the utterance identified by the utterance information, an utterance output unit which outputs the translated utterance information, a dialog history database, an utterance prediction unit which generates a first utterance prediction information based on the first utterance prediction information and the second utterance prediction information.
1. A dialog supporting apparatus which supports a present dialog between dialog participants, said dialog supporting apparatus comprising:
a dialog history database for storing a previous dialog of a first dialog participant, the previous dialog including a number of utterances;
an utterance prediction unit for (a) generating first utterance prediction information based on the utterances of the previous dialog stored in said dialog history database, (b) obtaining second utterance prediction information indicating a number of utterances of a previous dialog from another dialog supporting apparatus, and (c) predicting a next utterance in the present dialog of the first dialog participant using said dialog supporting apparatus based on the first utterance prediction information and the second utterance prediction information; and
a display unit for displaying the next utterance predicted by said utterance prediction unit, wherein:
said dialog history database stores a plurality of previous dialogs each including a number of utterances;
said utterance prediction unit extracts a previous dialog which is most similar to the present dialog from among the plurality of previous dialogs stored in said dialog history database, the first utterance prediction information being based on the utterances of the extracted previous dialog;
said utterance prediction unit generates a first prediction stack based on an assembly of successive utterances in the first utterance prediction information and on an assembly of successive utterances in the second utterance prediction information from the other dialog supporting apparatus; and
said utterance prediction unit judges whether or not an utterance is common to both the first and second utterance prediction information by performing at least one of a matching based on utterance information for identifying the utterance, a matching based on a surface expression and a matching based on a content word.