Personal audio systems are designed to deliver spatially separated regions of audio to individual listeners. This paper presents a method for improving the privacy of such systems. The level of a synthetic masking signal is optimised to provide specified levels of intelligibility in the bright and dark sound zones and reduce the potential for annoyance of listeners
in the dark zone by responding to changes in ambient noise.

Results from a simulated personal audio system indicate that less acoustic contrast is required to produce the same level of
privacy when artificial masking is included in the system design, compared with relying on the masking effect of background noise
alone. As privacy requirements become more challenging, the advantage gained by incorporating artificial masking increases.