PP1330 - The Effects of Talker Asymmetry on Speech Perception in Multi-Talker Listening Environments

Speech segregation can improve due to release from masking regarding talkers’ voice-gender differences and spatial separation. In this study, target-masker gender combinations were tested with four spatial conditions. Those conditions were repeated in both symmetric and asymmetric spatial configurations. Results revealed masking release by the spatial separation was increased up to 10 dB when the maskers were symmetrically separated from the target, and this spatial benefit was improved by 15 to 20 dB in the asymmetric target-masker condition. Results also revealed that masking release by voice-gender cue was not affected by the talker asymmetricity and fixed at 15 dB.

Summary: Rationale/Purpose    

Many previous studies have reported that speech segregation performance in multi-talker environments can be enhanced by following two major acoustic cues: 1) vocal-characteristic differences between talkers; 2) spatial separation between talkers. Here, the improvement they can provide for speech segregation 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 apparent source location of the target and competing speech. Our recent study demonstrated that the VGRM and SRM elicit an unequal perceptual weighting in the listener’s speech perception performance, and the magnitude of masking release is the same for the two cue types (e.g., equal perceptual weighting) at the ~ 104-Hz F0 difference and the ~ ±15° spatial separation between target and masker (Oh et al., 2021).The goal of this study was to further investigate how a listener’s perceptual weighting with voice and spatial cues is affected by talker asymmetry.
Methods

Twenty young adult normal-hearing listeners participated in the speech recognition in noise experiment. Speech sentences were drawn from the Coordinate Response Measure (CRM; Bolia et al., 2001) speech corpus spoken by four male (F₀ = 100 ± 7 Hz) and four female talkers (F₀ = 204 ± 12 Hz). Speech-on-speech masking performance was measured as the threshold target-to-masker ratio (TMR) 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 four spatial configurations (maskers co-located or 30°, 60°, and 90° spatially separated from the target) to manipulate the spatial separation cue. Target speech was fixed in front (0°), and two maskers were distributed asymmetrically toward the right or left or distributed symmetrically toward the right and left. The amounts of VGRM and SRM were computed from the TMR thresholds for each experimental condition.
Results/Conclusions

Average results showed that the amount of masking release by the spatial separation cue was increased up to 10 dB when the maskers were symmetrically separated from the target, and this spatial benefit was improved by 15 to 20 dB in the asymmetric target-masker condition. However, the results revealed that the amount of masking release by voice-gender difference cue was not affected by the talker asymmetry and was fixed at 15 dB. The findings of this study suggest that the relative reliance on voice gender difference and spatial separation cues can be influenced by target-masker asymmetry when two cues are interchangeable, and this effect solely depends on spatial-cue benefit. Further research will be conducted to examine the effects of talker asymmetry on speech-on-speech masking tasks for hearing-impaired listeners. This will allow for more inclusivity for future use to promote a hearing-impaired listener’s speech segregation performance in multi-talker situations.