| 摘要 |
PROBLEM TO BE SOLVED: To provide a navigation technology for effectively utilizing finite memory capacity according to the number of phrases to be recognized for each operation state. SOLUTION: A judging section 47 of a voice input control section 45 judges whether the number of phrases necessary for the next recognition exceeds the prescribed value according to the operation state. A recognition dictionary generating section 48, when the number of phrases is judged not to exceed the prescribed value, prepares phrase data of all the phrases necessary for the next recognition in a recognition dictionary 46 serving as the second storing means based on the first storing means, and when judged to exceed the prescribed value, prepares a part of the phrase data among each phrase necessary for the next recognition in the recognition dictionary 46 based on the first storing means. |
