| 摘要 |
PROBLEM TO BE SOLVED: To provide a transoral system which does not annoy a listener and capable of generating a binaural signal even though the listener largely moves.SOLUTION: A posture of the face of a listener is analyzed from face image information and face posture information is obtained, a head transfer function corresponding to the face posture information is convoluted to an audio signal so that binaural signals of a right channel and a left channel are generated. With the binaural signals and the face posture information as inputs, a transfer function from two speakers respectively selected from among three or more speakers corresponding to the face posture information to each of the right ear and the left ear of the listener is used to perform filter processing to remove space crosstalk component from the binaural signal. With the face posture information and the binaural signal output from a crosstalk processing section as inputs, a pair of speakers adjacent to each other are selected from among the three or more speakers and the binaural signal is output from each of the speakers. |
