The role of perceived spatial separation in the unmasking of speech

Spatial separation of speech and noise in an anechoic space creates a relea se from masking that often improves speech intelligibility. However, the ma sking release is severely reduced in reverberant spaces. This study investi gated whether the distinct and separate localization of speech and interfer ence provides any perceptual advantage that, due to the precedence effect, is not degraded by reflections. Listeners' identification of nonsense sente nces spoken by a female talker was measured in the presence of either speec h-spectrum noise or other sentences spoken by a second female talker. Targe t and interference stimuli were presented in an anechoic chamber from louds peakers directly in front and 60 degrees to the right in single-source and precedence-effect (lead-lag) conditions. For speech-spectrum noise, the spa tial separation advantage for speech recognition (8 dB) was predictable fro m articulation index computations based on measured release from masking fo r narrow-band stimuli. The spatial separation advantage was only 1 dB in th e lead-lag condition, despite the fact that a large perceptual separation w as produced by the precedence effect. For the female talker interference, a much larger advantage occurred, apparently because informational masking w as reduced by differences in perceived locations of target and interference . (C) 1999 Acoustical Society of America. [S0001-4966(99)01412-5].