Information Technology Reference
In-Depth Information
(a) Results for utterances 1 and 2.
(b) Results for utterances 2 and 3.
(c) Results for utterances 3 and 4.
Fig. 2.
Bar diagrams presenting results of the topic tracking process for our example dialog (see
2.1). For clarity reasons, the display adapts automatically and only shows those bars representing
active topics.
topic awareness in artificial agents our presented topic detection model needs to be
embedded into the agent's underlying system architecture. More precisely, the agent's
conversational behavior has to be adapted by means of the gained topic information to
enable coherent dialogs between human and artificial interlocutors. In this section we
propose our approach for improving the conversational abilities of the ECA Max by
integrating topic information into the agent's existing dialog system.
4.1
The Conversational Agent Max
Max
[15] is a virtual character acting as a conversational computer interface that
allows for face-to-face dialogs with humans in German language. By means of keyboard-
based, textual inputs human users are able to communicate with the agent. Max is capa-
ble of responding to these inputs with spoken language realized by a synthesized voice.
