Spatial Release from Masking for Small Spatial Separations Using Simulated Cochlear Implant Speech

Abstract

Background: Spatial release from masking (SRM) is the improvement in speech intelligibility when the masking signals are spatially separated from the target signal. Young, normal- hearing listeners have a robust auditory sys-tem that is capable of using the binaural cues even with a very small spatial separation between the target and the maskers. Prior studies exploring SRM through simulated cochlear implant (CI) speech have been completed using substantial spatial separations, exceeding 45° between the target signal and masking signals. Nevertheless, in re-al-world conversational scenarios, the spatial separation between the target and the maskers may be considerably less than what has been previously investigated. This study presents SRM data utilizing simulated CI speech with young, normal-hearing listeners, focusing on smaller but realistic spatial separations between the target and the maskers. Methods: Twenty-five young, normal-hearing listeners participated in this study. Speech identification thresholds, the target-to-masker ratio required to accurately identify 50% of the target words, were measured for both natural speech and simulated CI speech. Results: The results revealed that young, normal-hearing listeners had significantly higher speech identification thresholds when presented with simulated CI speech in comparison to natural speech. Furthermore, the amount of SRM was found to be greater for natural speech than for the simulated CI speech. Conclusions: The data suggests that young normal-hearing individuals are capable of utilizing the interaural level difference cues in the simulated cochlear implant signal to achieve masking release at reduced spatial separations between the target and the maskers, highlighting the auditory system’s capability to extract these interaural cues even in the presence of degraded speech signals.