Release from masking through spatial separation in distance

It is widely accepted that speech intelligibility improves as a speech signal and interfering masker are separated spatially in azimuth. How-ever, little attention has been given to the effect of spatial separation in distance. In the current study speech reception thresholds (SRTs) were investigated using two different measures, the Listening in Spatialized Noise – Sentences Test (LiSN-S) and the Coordinate Response Measure (CRM). Both corpuses are characterized by providing a large degree of informational masking. To compare the results with pure energetic masking, the SRT was additionally measured in a reference condition using speech-shaped noise (SSN). Two different setups were realized. In a first setup the target was presented at a distance of 0.5m from the center of the listener‘s head and the interferer at a distance of 0.5 m, 1 m, 2 m or 10 m. In a second setup the interferer‘s location was fixed and the target‘s location was varied accordingly. Spatial synthesis was realized through convolution with binaural room impulse responses (BRIRs) recorded in an auditorium. In order to compensate for the distance dependent change in overall intensity, the total signal power was equalized and the long-term frequency spectra of the interferers were adjusted to the average target spectrum. In the case of the SSN interferer the SRT was unaffected by changes in distance. However, for the speech interferer the results revealed a substantial release from masking as the target and interferer were separated in distance. This effect was consistent for both the LiSN-S and CRM measure as well as for both target-masker setups. The strongest release from masking effect of about 10 dB was observed for the CRM corpus when the target talker was at 0.5 m and the distractor was moved from 0.5m to 10 m. In this configuration the SRT was similar to the one observed for the SSN interferer. This study suggests that distance related cues play a significant role when listening in complex environments and that the ability to use such cues could relate directly to how normal hearing and hearing impaired listeners function in complex scenes.