Speech-on-Speech Masking by Voice-Gender Differences and Spatial Separation Cues: Normal Hearing Versus Simulated Single-Sided Deafness

PURPOSE: Many previous studies on normal-hearing (NH) listeners have reported that the ability to use binaural (spatial) and monaural (voice-gender) cues can enhance speech segregation performance in multitalker environments, referred to as "release from masking." This study aimed to investigate how simulated single-sided deafness (SSD) affects masking release performance, particularly in relation to voice-gender differences and spatial separation between talkers. METHOD: Twenty NH listeners (aged 20-26 years) measured speech recognition thresholds (SRTs) in binaural and simulated SSD listening conditions. Target speech was fixed in front (0°), and two competing maskers were spatially separated from -90° to +90° with 30° increments. The measurements were repeated for same-gender and different-gender target-masker combinations. The amounts of masking release by voice and spatial cues were computed from SRTs. RESULTS: Binaural listeners showed that the masking release occurred symmetrically with interchanging perceptual weighting between voice and spatial cues. In the simulated SSD, the masking release was primarily mediated by voice-gender difference benefits. However, interaction effects demonstrated equal perceptual weighting of both cues when the maskers were separated by 60° and presented ipsilateral to the simulated deafened ear. CONCLUSIONS: The loss of binaural input compels the auditory system to shift reliance from cue integration to strategic prioritization. This study suggests that listeners with simulated SSD could maximize speech segregation by leveraging voice-gender difference cue and the passive head-shadow effect when noise sources are positioned approximately 60° toward the deafened ear.