摘要 |
PROBLEM TO BE SOLVED: To discriminate a user as an 'inexperienced' user when hesitation and an unnecessary word exist in a response even though the reaction time of the response is same as the one made by an experienced use who is skilled in operation. SOLUTION: A voice recognition section 1, which receives voice uttered by a user, recognizes the order of uttering of the words and the phrases that are beforehand registered in a voice recognition dictionary section 2. An unnecessary word detecting section 3 checks whether an unnecessary word is included in the recognition result of the section 1 or not. When an unnecessary word is included, the positional relationship between the unnecessary word and an objective word in the recognition result is checked. A skill estimating section 4 estimates the skill of the operations of the voice responding device of the user based on the checked result obtained by the section 3. A conversation flow control section 5 takes out the guidance included in the conversation flow corresponding to the skill estimated by the section 4 among the conversation flow beforehand stored in a conversation flow storage section 6. A guidance output section 7 sends the guidance to the user.
|