Binaural unmasking cannot explain spatial release from masking in multispeaker situations

Evidence

The stable spatial release from masking over the range of SNR criteria tested is one of the most important outcomes of this study. It suggests that recovery of sub-threshold speech information by binaural cues does not play a role in the spatial release from masking in a multispeaker situation. Instead, masking release appeared to be sustained even when only high-SNR components of target speech were left in the signal. These supra-threshold components should be detectable monaurally in both the collocated and spatially separated conditions and thus their detection should not rely on a binaural unmasking mechanism as is the case for BMLDs. An explanation for this spatial release from masking can be provided by directional cues of the stimuli that support a better allocation of speech components to the source of interest. — Dijk, Pim van, Physiology, Psychoacoustics and…, 1661

The stable spatial release from masking over the range of SNR criteria tested is one of the most important outcomes of this study. It suggests that recovery of sub-threshold speech information by binaural cues does not play a role in the spatial release from masking in a multispeaker situation. Instead, masking release appeared to be sustained even when only high-SNR components of target speech were left in the signal. These supra-threshold components should be detectable monaurally in both the collocated and spatially separated conditions and thus their detection should not rely on a binaural unmasking mechanism as is the case for BMLDs. — Dijk, Pim van, Physiology, Psychoacoustics and…, 1670

The stable spatial release from masking over the range of SNR criteria tested is one of the most important outcomes of this study. It suggests that recovery of sub-threshold speech information by binaural cues does not play a role in the spatial release from masking in a multispeaker situation. Instead, masking release appeared to be sustained even when only high-SNR components of target speech were left in the signal. These supra-threshold components should be detectable monaurally in both the collocated and spatially separated conditions and thus their detection should not rely on a binaural unmasking mechanism as is the case for BMLDs. An explanation for this spatial release from masking can be provided by directional cues of the stimuli that support a better allocation of speech components to the source of interest. The fact that speech sources originate from different spatial positions can eliminate some confusion that arises from the simultaneous presence of multiple speech sources (Durlach et al. — Dijk, Pim van, Physiology, Psychoacoustics and…, 1619

When listening to speech in a noisy background, higher speech intelligibility is measured for spatially separated speech and interfering sources, as compared to collocated sources. The benefit of this spatial separation, known as spatial release from masking, has been attributed to binaural processing enabling a better signal detection, as can be measured by binaural masking level differences (BMLDs). These BMLDs describe the difference in thresholds for tone-in-noise detection when both the target tone and masking noise are presented interaurally in phase, in contrast to a reversed phase of the tone in one ear. In the latter case thresholds are significantly lower. Similar interaural relations of the stimuli can be achieved by presenting the noise from a spatial position directly in front of the listener and the signal from a different spatial location. Therefore a connection was hypothesized between the spatial benefit in speech intelligibility and the tone detection advantage that had been measured as BMLD. — Dijk, Pim van, Physiology, Psychoacoustics and…, 1619

Categorization

• XREF: Contrasts the classic BMLD (binaural masking level difference) literature with the modern cocktail-party spatial release picture; the mechanism invoked here is source segregation/allocation via directional cues, not binaural detection advantage. — connects to: Binaural intelligibility level difference measures speech unmasking benefit — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 1513
• QUESTION: Could the result depend on the restricted vocabulary task and learned words, masking an effect that would appear with open-set materials?
• DEEPTHINKING: Undermines a textbook mechanism for cocktail-party listening in real multi-speaker situations; the stable masking release has implications for exactly what binaural hearing contributes in everyday scenes.
• DEFINE: Defines spatial release from masking and BMLDs with concrete in-phase/reversed-phase threshold contrast.
• corroborated: 4 independent passages support this claim — Dijk, Pim van, Physiology, Psychoacoustics and…, 1661; Dijk, Pim van, Physiology, Psychoacoustics and…, 1670; Dijk, Pim van, Physiology, Psychoacoustics and…, 1619