Segregation of Multiple Talkers in the Vertical Plane: Implications for the Design of a Multiple Talker Display

Abstract

Three experiments were conducted to evaluate the effect of spatial separation of multiple talkers in the vertical plane on speech intelligibility. The first experiment demonstrated a release from masking due to separation in the median plane, and that this release was not due to the presence of residual interaural time differences (ITDs). The second experiment showed that this release corresponded to an increase in signal level of 1.3 dB. The third experiment demonstrated that the increase in intelligibility due to separation in elevation and that due to separation in azimuth were not additive. Results are discussed in terms of their implications for the design of spatial audio displays.