Arbogast Tanya L, Mason Christine R, Kidd Gerald
Hearing Research Center and Communication Disorders, Boston University, 635 Commonwealth Avenue, Boston, Massachusetts 02215, USA.
J Acoust Soc Am. 2005 Apr;117(4 Pt 1):2169-80. doi: 10.1121/1.1861598.
The ability to understand speech in a multi-source environment containing informational masking may depend on the perceptual arrangement of signal and masker objects in space. In normal-hearing listeners, Arbogast et al. [J. Acoust. Soc. Am. 112, 2086-2098 (2002)] found an 18-dB spatial release from a primarily informational masker, compared to 7 dB for a primarily energetic masker. This article extends the earlier work to include the study of listeners with sensorineural hearing loss. Listeners performed closed-set speech recognition in two spatial conditions: 0 degrees and 90 degrees separation between signal and masker. Three maskers were tested: (1) the different-band sentence masker was designed to be primarily informational; (2) the different-band noise masker was a control for the different-band sentence; and (3) the same-band noise masker was designed to be primarily energetic. The spatial release from the different-band sentence was larger than for the other maskers, but was smaller (10 dB) for the hearing-impaired group than for the normal-hearing group (15 dB). The smaller benefit for the hearing-impaired listeners can be partially explained by masker sensation level. However, the results suggest that hearing-impaired listeners can use the perceptual effect of spatial separation to improve speech recognition in the presence of a primarily informational masker.
在包含信息掩蔽的多源环境中理解语音的能力可能取决于信号和掩蔽对象在空间中的感知排列。在听力正常的听众中，阿博加斯特等人[《美国声学学会杂志》112, 2086 - 2098 (2002)]发现，相对于主要是能量掩蔽的情况为7分贝，从主要是信息掩蔽中可获得18分贝的空间解脱。本文将早期的工作进行了扩展，纳入了对感音神经性听力损失听众的研究。听众在两种空间条件下进行闭集语音识别：信号与掩蔽之间的分离度为0度和90度。测试了三种掩蔽：(1)不同频段句子掩蔽设计为主要是信息性的；(2)不同频段噪声掩蔽作为不同频段句子的对照；(3)同频段噪声掩蔽设计为主要是能量性的。来自不同频段句子的空间解脱比其他掩蔽更大，但听力受损组（10分贝）比听力正常组（15分贝）要小。掩蔽感觉水平可部分解释听力受损听众获益较小的原因。然而，结果表明，听力受损听众可以利用空间分离的感知效应来改善在主要是信息掩蔽存在时的语音识别。
