The significance of representing realistic torso orientation relative to the head in HRTF is studied in this work. When the head turns, the torso typically remains stationary, resulting in an variation of the angle between the head and torso. Rotated torso is compared to a torso remaining fixed to the head orientation, typical to HRTF data. Actual head locations relative to torso are found for 195 persons using their 3D models. The effect of the head position in HRTF is studied using 3D model of Kemar head and torso simulator geometry by translating the head in up-down and forward-backward directions and rotating it. The closest matching persons are found to the Kemar cases with translated head positions, and for these persons the heads are rotated but not translated in 3D models. The torso effect in the modified Kemar shows good agreement with the actual persons. Head position relative to shoulders effects the torso signature in HRTF. Front-back location of the head has the strongest influence in the HRTF. The spectral difference between the fixed and rotated torso spectra can exceed 1 dB limit in all sound arrival azimuth directions when the torso rotation exceeds 10 degrees. The spectral difference decreases with increasing sound source elevation. Torso rotation can be audible for any significant torso rotation.
(See document for exact affiliation information.)
https://aes2.org/publications/elibrary-page/?id=22213
(1105KB)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
