著者
高根 昭一 安倍 幸治 渡邉 貫治 西口 正之
出版者
特定非営利活動法人 日本バーチャルリアリティ学会
雑誌
日本バーチャルリアリティ学会論文誌 (ISSN:1344011X)
巻号頁・発行日
vol.26, no.4, pp.266-276, 2021-12-24 (Released:2021-12-24)
参考文献数
37

Head-Related Transfer Function (HRTF) is an acoustic transfer function corresponding to sound transmission from a sound source to one’s ear in a free field. Its property depends on sound source position and difference in shapes of individual ear, head and torso and so on. Acquiring the HRTFs of a certain subject for multiple source positions based on measurement demands that the subject stationarily exists in the measurement system for a long time. It brings about serious physical burden to the subject, and causes some variance in the measured HRTFs. In order to resolve this difficulty, a 3D shape of the subject’s head and ears was scanned by using a 3D scanner and a digital camera, and the 3D data of the subject’s head was constructed. This is called “head model” in this paper. Then the head model was printed out by using a 3D printer, and the HRTFs of the subject were measured locating the head model instead of letting the subject be in the measurement system. As a fundamental investigation, a HATS (Head-and-Torso Simulator) was selected as a subject, and its HRTFs and those of its head model were measured in horizontal plane. As a result, the HRTFs measured for the head model of the HATS are an appropriate approximation with those of the HATS itself in the frequency range under 10 kHz, showing the possibility that the head model of the subject can be a substitute for the measurement of the subject’s HRTFs.