Objective and subjective analysis of effects of listening angle on crosstalk cancellation in spatial sound reproduction

Abstract

A comprehensive study was conducted to explore the effects of listening angle on crosstalk cancellation in spatial sound reproduction using two-channel stereo systems. The intention is to establish a sustainable configuration of crosstalk cancellation system (CCS) that best reconciles the separation performance and the robustness against lateral head movement, not only in theory but also in practice. Although crosstalk can in principle be suppressed using multichannel inverse filters, the CCS does not lend itself very well to practical application owing to the fact that the sweet spot is being so small. Among the parameters of loudspeaker deployment, span angle is a crucial factor that has a profound impact on the separation performance and sweet spot robustness achievable by the CCS. This paper seeks to pinpoint, from a more comprehensive perspective, the optimal listening angle that best reconciles the robustness and performance of the CCS. Two kinds of definitions of sweet spot are employed for assessment of robustness. In addition to the point source model, head related transfer functions (HRTF) are employed as the plant models in the simulation to emulate more practical localization scenarios such as the high-frequency head shadowing effect. Three span angles including 10, 60, and 120 deg are then compared via objective and subjective experiments. The Friedman test is applied to analyze the data of subjective experiments. The results indicate that not only the CCS performance but also the panning effect and head shadowing will dictate the overall performance and robustness. The 120-deg arrangement performs comparably well as the standard 60-deg arrangement, but is much better than the 10-deg arrangement.

Objective evaluation of the sweet spot size in spatial sound reproduction using elevated loudspeakers

“…6,13 In general, closely spaced loudspeakers ͑ s = 12°͒ result in a wider effective area than the typical stereo setup ͑ s = 60°͒. We can also see that the region of ringing frequencies shifts to higher frequencies as the loudspeakers get closer to each other.…”

“…6, in which the absolute sweet spot increased with wider span angles. In that study, the channel separation index was calculated for different bandwidths in which the lower bound was 100 Hz and the regularization was set constant with respect to frequency.…”

“…To get a better overview of the results, we made use of the two sweet spot definitions described by Bai et al 6 The first one is the absolute sweet spot, which is defined as two times the maximum leftward displacement from the nominal center position that results in a channel separation index below Ϫ12 dB. The second one is the relative sweet spot, which is two times the maximum leftward displacement from the nominal center position that results in a channel separation index degradation of 12 dB with respect to the nominal center position.…”

“…6,13 One of the reasons for obtaining a wider sweet spot with closely spaced loudspeakers is that the variations of the distance between the listener and the loudspeakers are reduced when the loudspeakers are set close together. 10 If we now place the loudspeakers above the head, these distance variations would be even smaller for certain kind of movements, such as head rotations.…”

“…Some studies have also proposed different span angles and loudspeaker arrangements in order to improve the performance of the binaural reproduction by widening the sweet spot. [6][7][8][9][10] However, most of them have mainly focused on loudspeakers placed in the horizontal plane.…”