PP513 - Speech-on-Speech Masking by Voice-Gender Difference and Spatial Separation Between Talkers in Listeners with Simulated Single-Sided Deafness

Abstract:
Normal hearing (NH) listeners can enhance speech segregation abilities using talkers’ voice-gender difference and spatial separation cues. However, listeners with single-sided deafness (SSD) have poor segregation performance due to limited access to binaural cues. This study investigated how SSD listeners perform speech-on-speech masking by voice and spatial cues. Results revealed that SSD listeners heavily rely on the voice-gender difference cue. However, presenting the maskers on deaf ears and keeping them 60° away from the target increased the spatial benefit. Further, the benefits by talker’s voice-gender difference and spatial separation were equally weighted at 60° target-masker separation toward the deaf ear.

Summary:
Rationale/Purpose
Many previous studies with normal hearing (NH) listeners have reported that speech segregation performance in multi-talker environments can be enhanced by following two major acoustic cues: 1) talkers’ vocal-characteristic differences; 2) talkers’ spatial separation. Here, the improvement is referred to as “release from masking”. Masking release is greater when target and masker talkers are different genders than when they are the same gender, which is referred to as “Voice Gender Release from Masking” (VGRM). In addition, “Spatial Release from Masking” (SRM) is facilitated by increasing differences in the location of the target and competing speech. Our recent results show that the VGRM and SRM elicit unequal perceptual weighting in the NH listener’s speech perception performance (Oh et al., 2021). However, listeners with single-sided deafness (SSD) have poor masking release performance due to limited access to binaural cues. This study investigated how SSD listeners perform speech-on-speech masking by voice and spatial cues and how the perceptual weighting between VGRM and SRM differs from that of NH listeners.

Methods
Twenty young adult NH listeners participated in the speech recognition in noise experiment. Speech-on-speech masking was performed in binaural and monaural (i.e., simulated SSD) listening conditions and measured speech recognition thresholds needed to understand a target talker in the presence of either same- or different-gender masker talkers to manipulate the voice-gender difference cue. These target-masker gender combinations were also tested with seven spatial configurations (-90°, -60°, -30°, 0°, +30°, +60°, and +90° separated from the target) to manipulate the spatial separation cue. Target speech was fixed at 0°, and two maskers were presented to the left or right. The amounts of VGRM and SRM were computed from the thresholds for each experimental condition.

Results
Average results showed that, for the binaural listening condition, the VGRM was maximized at 15dB when the target and maskers were co-located, and the VGRM decreased by 12dB as the maskers were separated from the target. On the contrary, the SRM was increased up to 18dB with target-masker separation greater than 30°. However, the results for the monaural (simulated SSD) listening condition showed that the VGRMs (8-12dB) were higher than SRMs (-5-8dB) in all target-masker spatial separations. The results also showed the same VGRM and SRM (8dB) when the maskers were presented on deaf ears at 60° away from the target. The results by a linear mixed model (LMM) analysis showed that the VGRM and SRM were significantly different in both binaural and monaural (SSD) listening conditions at all target-masker spatial separations, except for the +60 spatial separation.

Conclusion
The findings of this study suggest that the relative reliance on voice gender difference and spatial separation cues is different between NH and SSD listeners when two cues are interchangeable. Further research will be conducted to examine the effects of the degree of asymmetric hearing loss on speech-on-speech masking tasks for hearing-impaired listeners. This will allow for more inclusivity for future use to promote speech segregation performance in multi-talker situations for hearing-impaired listeners with various types of hearing loss.