摘要 |
PROBLEM TO BE SOLVED: To prevent voice from being recognized with low accuracy when a speaking person is not near a microphone. SOLUTION: The content-of-speech recognition device 10 includes a microphone 16 for collecting voice, an imaging device 18 for capturing an image of the person who speaks toward the microphone 16, a voice recognition function unit 132 which recognizes voice based on the collected voice, and a recognition/learning determination section 142 which restrains voice recognition of the voice recognition function unit 132 when the captured image does not include an image of the person which shows at least a part of the person. COPYRIGHT: (C)2011,JPO&INPIT |