MODELS OF HEARING LOSS

听力损失模型

• 批准号: 2125534
• 负责人: CHRISTOPHER W TURNER
• 金额: $ 14.87万
• 依托单位: SYRACUSE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 1989
• 资助国家: 美国
• 起止时间: 1989-05-01 至 1998-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/2125534
• 关键词: auditory discrimination; auditory threshold; cues; disease /disorder model; hearing aids; hearing tests; human subject; loudness; model design /development; perceptual maskings; psychoacoustics; sensorineural hearing loss; sound frequency; speech recognition; speech synthesizers

This application describes experiments on the (mis)perception of speech cues by individuals with sensorineural hearing loss. The general approach is to use speech and speech-like stimuli, as opposed to typical psychoacoustic stimuli. The experiments identify the types of speech segments and cues that are recognized incorrectly by hearing-impaired listeners. A measure of audibility based upon the short-term spectrum of speech sounds is used to separate the contribution of reduced audibility (i.e., errors due to speech cues presented below listener's thresholds) from the contribution of suprathreshold factors. Extensive studies of the levels of speech stimuli and audibility of individual speech sounds are also described. Additional experiments are proposed to identify the specific types of suprathreshold speech cures (described in physical terms, i.e., formant transitions, speech envelope cues, etc.) that present difficulties for the hearing-impaired individual. The objective is to search for common characteristics of these "problem speech cues" in order to lay the ground work for a comprehensive approach to explaining suprathreshold speech errors. Follow-up experiments will investigate the effects of modifying these speech cures upon their ability to be used by hearing-impaired subjects. Strategies including compression, and recording of formant transitions will be tested. The effect of long-term training upon listeners' abilities to use these modified cues will also be studied. The overall objective is to identify the types of speech cues that will require more than amplification to maximize speech recognition by patients and to investigate the bases for these results. The use of speech and speech-like signals is a unique approach in studying hearing loss, in that the specific problems of hearing-impaired subjects to use various speech cues can be studied directly. This knowledge has application to both improved use of existing hearing aid technologies, and also to improving future hearing aids.

这个应用程序描述了(错觉)语音感知的实验。 由感音神经性耳聋患者提供的提示。一般的方法 是使用语音和类似语音的刺激，而不是典型的 心理声学刺激。这些实验识别语音的类型 听力受损人士不能正确识别的片段和提示 听众。一种基于短时频谱的可听性度量 语音被用来分离降低的可听度的贡献 (即，由于语音提示低于听者的阈值而产生的错误) 来自阈值以上因素的贡献。对人类的广泛研究 语音刺激的级别和单个语音的可听性是 还描述了。 建议进行额外的实验，以确定特定类型的 阈值以上语音治疗(用物理术语描述，即共振峰 过渡、语音包络提示等)这给美国政府带来了困难 听力受损的人。目标是寻找共同之处 这些“问题语音提示”的特点，以期奠定基础 致力于用一种全面的方法来解释阈值以上的言语 错误。后续实验将调查改装后的效果 这些语言治疗听力受损的人使用它们的能力 研究对象。包括压缩和记录共振峰的策略 将对过渡进行测试。长期训练对人体健康的影响 听者使用这些修改过的线索的能力也将被研究。 总体目标是确定语音提示的类型 需要的不仅仅是放大功能，才能最大限度地提高患者的语音识别能力 并调查这些结果的基础。语言和语言的使用 类语音信号是研究听力损失的一种独特的方法，因为 听障人士使用多种语言的具体问题 线索可以直接研究。这一知识对两者都适用。 改进现有助听器技术的使用，并改进 未来的助听器。