摘要 |
PROBLEM TO BE SOLVED: To make possible reducing erroneous recognition even when a large capacity vocabulary data storage part is provided. SOLUTION: A microphone 9 and a talk switch 10 are connected to a speech recognition system 8, and a voice signal inputted to the microphone 9 is amplified by an amplifier 11 to be inputted to a recognition processing part 12. The attribute vocabulary data added with attributes are stored in a vocabulary data storage part 13 at every its attribute. The recognition processing part 12 reds out the attribute vocabulary data having the attribute inputted to the microphone 9 by a user, and makes them objective vocabularies of voice recognition, and thereafter, recognizes the voice inputted to the microphone 9 form the attribute vocabulary data. Further, and on-vehicle navigator 1 is operated by an operation processing part 5 based on the vocabulary recognized by the speech recognition system 8. |