| 摘要 |
<p>A video-image analyzing section (121) extracts at least one type of feature information from an image, such as the face of a viewer/listener, acquired from video-image information received from a video-image input section (11), generates feature extraction data (131) for each type of extracted feature information by numerically expressing each of the features contained in each feature information, and stores the feature extraction data in a storage device (13). A language determining section (122) determines the language the viewer/listener is most likely to use on the basis of: the values of the features in the feature extraction data (131); and language usability data (132) that indicates, for each of said features in the feature extraction data, the degree of likelihood that each of a plurality of languages is used. A display/audio control section (123) displays, on a display section (14), and aurally outputs, with an audio output section (15), language data of the language determined by the language determining section (122) that the viewer/listener is most likely to use from among a plurality of pieces of language data (133).</p> |
