Rählmann S, Meister H
Jean-Uhrmacher-Institut für klinische HNO-Forschung, Universität zu Köln, Geibelstr. 29-31, 50931, Köln, Deutschland.
HNO. 2017 Mar;65(3):228-236. doi: 10.1007/s00106-016-0303-y.
In communication situations with multiple speakers, speech recognition is adversely affected by energetic masking (EM) and informational masking (IM). IM characterizes masking effects caused by irrelevant information from competing speakers. This work investigates an approach to assess IM based on the Oldenburg Sentence Test (OLSa). Furthermore, the influence of interaural time differences (ITD) and aging effects on IM are considered.
IM was measured by superimposing two sentences from the OLSa. The beginning of the target sentence was indicated by the keyword "Stefan". To segregate between target and masker sentences, ITDs from 50 to 400 µs were included. The participants were asked to selectively attend to the target sentence and repeat back the spoken words. Potential factors associated with speech recognition results were assessed by an auditory measure of temporal fine structure and a neuropsychological profile. 16 normal-hearing listeners between 18 and 77 years of age participated in the study.
Despite the clinically normal-hearing participants, the analysis showed a significant relationship between speech recognition outcome and pure-tone thresholds. All participants benefited from small ITDs between the target and masker sentence with regard to the unmasking of IM. The magnitude of unmasking could not be explained by any of the factors assessed in this study. Error analysis and the comparison to the literature reveal that the OLSa could be a useful tool to assess IM. Also in line with the current literature is the relationship between speech recognition outcome and pure-tone thresholds, as well as the strong effect of ITDs on the release from IM.
Speech audiometric assessment of IM is of high relevance with regard to everyday communication situations. Due to its structure, the OLSa seems to be a useful tool for determining IM.
在多说话者的交流情境中，语音识别会受到能量掩蔽（EM）和信息掩蔽（IM）的不利影响。IM表征了来自竞争说话者的无关信息所引起的掩蔽效应。本研究探讨了一种基于奥尔登堡句子测试（OLSa）来评估IM的方法。此外，还考虑了双耳时间差（ITD）和衰老效应对IM的影响。
通过叠加来自OLSa的两个句子来测量IM。目标句子的开头由关键词“Stefan”指示。为了区分目标句子和掩蔽句子，纳入了50至400微秒的ITD。要求参与者选择性地关注目标句子并重复说出听到的单词。通过对时间精细结构的听觉测量和神经心理学概况来评估与语音识别结果相关的潜在因素。16名年龄在18至77岁之间的听力正常的听众参与了这项研究。
尽管参与者临床听力正常，但分析表明语音识别结果与纯音阈值之间存在显著关系。所有参与者在目标句子和掩蔽句子之间存在小的ITD时，在解除IM方面都受益。本研究评估的任何因素都无法解释解除掩蔽的程度。错误分析以及与文献的比较表明，OLSa可能是评估IM的有用工具。语音识别结果与纯音阈值之间的关系以及ITD对解除IM的强烈影响也与当前文献一致。
在日常交流情境中，对IM进行言语听力评估具有高度相关性。由于其结构，OLSa似乎是确定IM的有用工具。
