A COMPARISON OF THE PERFORMANCE OF HRTF MODELS IN INVERSE FILTER DESIGN FOR CROSSTALK CANCELLATION
Jacob Hollebon, E. Hamdan, Filippo Maria Fazi · 2023
Crosstalk cancellation can be used to reproduce binaural audio over loudspeakers without the use of headphones, by using inverse filtering. When designing the inverse filters, a matrix of head-related transfer functions (HRTFs) is modelled and inverted. Often in crosstalk cancellation, a free-field shadowless head model is assumed. For perfect crosstalk cancellation, the modelled HRTFs that are inverted should be identical to the physical HRTFs of the reproduction system. Mismatches between the modelled HRTFs and the physical HRTFs result in errors in the reproduced binaural signals. A range of HRTF models are reviewed and their performance in a standard crosstalk cancellation problem assessed using numerical simulations. The benefits and drawbacks of the various models are discussed and the limitations of the commonly used shadowless head model shown.