For the reproduction of binaural signals via loudspeakers, crosstalk cancellation systems are necessary. To compute the crosstalk cancellation filters, the transfer functions between loudspeakers and ears must be given. If the listener moves the filters are usually updated based on a model or previously measured transfer functions. We propose a novel architecture: It is suggested to place microphones close to the listener’s ears to continuously estimate the true transfer functions and use those to adapt the crosstalk cancellation filters. A fast frequency-domain state-space approach is employed for multichannel system tracking. For simulations of slow listener rotations it is demonstrated by objective and subjective means that the proposed system successfully attenuates crosstalk of the direct sound components.
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 and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.