Suppr超能文献

在透明模式下使用可穿戴式听力设备进行声音定位。

Sound Localization with Hearables in Transparency Mode.

作者信息

Ausili Sebastian A, Erthal Nathan, Bennett Christopher, Snapp Hillary A

机构信息

Department of Otolaryngology, University of Miami, 1120 NW 14th Street, 5th Floor, Miami, FL 33136, USA.

Department of Music Engineering, University of Miami, 1314 Miller Dr, Coral Gables, FL 33146, USA.

出版信息

Audiol Res. 2025 Apr 25;15(3):48. doi: 10.3390/audiolres15030048.

Abstract

Transparency mode in hearables aims to maintain environmental awareness by transmitting external sounds through built-in microphones and speakers. While technical assessments have documented acoustic alterations in these devices, their impact on spatial hearing abilities under realistic listening conditions remains unclear. This study aimed to evaluate how transparency mode affects sound localization performance with and without background noise. Ten normal-hearing adults completed sound localization tasks across azimuth (±90°) and elevation (±30°) with and without background noise. Performance was assessed with and without AirPods Pro in transparency mode. Sound localization performance was evaluated through linear regression analysis and mean absolute errors. Head-Related Transfer Function measurements quantified changes in binaural and spectral cues. While interaural time differences were largely preserved, transparency mode introduced systematic alterations in level differences (up to 8 dB) and eliminated spectral cues above 5 kHz. These modifications resulted in increased localization errors, particularly for elevation perception and in noise. Mean absolute errors increased from 6.81° to 19.6° in azimuth and from 6.79° to 19.4° in elevation without background noise, with further degradation at lower SNRs ( < 0.05). Response times were affected by background noise ( < 0.001) but not by device use. Current transparency mode implementation significantly compromises spatial hearing abilities, particularly in noisy environments typical of everyday listening situations. These findings highlight the need for technological improvements in maintaining natural spatial cues through transparency mode, as current limitations may impact user safety and communication in real-world environments.

摘要

可听设备中的通透模式旨在通过内置麦克风和扬声器传输外部声音来保持对环境的感知。虽然技术评估已经记录了这些设备中的声学变化,但它们在实际聆听条件下对空间听觉能力的影响仍不明确。本研究旨在评估通透模式在有无背景噪声的情况下如何影响声音定位性能。十名听力正常的成年人在有无背景噪声的情况下完成了在方位角(±90°)和仰角(±30°)上的声音定位任务。在通透模式下使用和不使用AirPods Pro评估性能。通过线性回归分析和平均绝对误差评估声音定位性能。头部相关传递函数测量量化了双耳和频谱线索的变化。虽然双耳时间差在很大程度上得以保留,但通透模式在声级差方面引入了系统性变化(高达8分贝),并消除了5千赫以上的频谱线索。这些变化导致定位误差增加,特别是在仰角感知和有噪声的情况下。在无背景噪声时,方位角的平均绝对误差从6.81°增加到19.6°,仰角从6.79°增加到19.4°,在较低信噪比(<0.05)时进一步恶化。反应时间受背景噪声影响(<0.001),但不受设备使用影响。当前的通透模式实施显著损害了空间听觉能力,特别是在日常聆听情况中典型的嘈杂环境中。这些发现凸显了通过通透模式保持自然空间线索方面技术改进的必要性,因为当前的局限性可能会影响现实环境中的用户安全和通信。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验