摘要 |
<p><P>PROBLEM TO BE SOLVED: To allow even a person having no professional knowledge to make conversation contents based on more accurate speech recognition, by creating a language model based on a conversation scenario for making conversation contents according to conversation flow which is prepared beforehand, while responding to user's utterance. <P>SOLUTION: This interaction scenario editing device which creates the conversation scenario for an automatic conversation system including: a conversation device which creates an input sentence through speech recognition of the user's utterance, and which requests a reply sentence according to the input sentence; and an conversation server which determines the reply sentence based on the conversation scenario, when the reply sentence is requested from the conversation device, and the reply sentence is transmitted to the conversation device and output for the user. The editing device includes a language model creating section for creating the language model to be used for the speech recognition based on the conversation scenario. <P>COPYRIGHT: (C)2010,JPO&INPIT</p> |