Department of Audiology & Speech-Language Pathology, University of North Texas, Denton, Texas.
Hearts for Hearing Foundation, Oklahoma City, Oklahoma.
J Am Acad Audiol. 2022 Mar;33(3):142-148. doi: 10.1055/s-0041-1740517. Epub 2022 Oct 10.
Cochlear implant (CI) recipients often experience speech recognition difficulty in noise in small group settings with multiple talkers. In traditional remote microphones systems, one talker wears a remote microphone that wirelessly delivers speech to the CI processor. This system will not transmit signals from multiple talkers in a small group. However, remote microphone systems with multiple microphones allowing for adaptive beamforming may be beneficial for small group situations with multiple talkers. Specifically, a remote microphone with an adaptive multiple-microphone beamformer may be placed in the center of the small group, and the beam (i.e., polar lobe) may be automatically steered toward the direction associated with the most favorable speech-to-noise ratio. The signal from the remote microphone can then be wirelessly delivered to the CI sound processor. Alternately, each of the talkers in a small group may use a remote microphone that is part of a multi-talker network that wirelessly delivers the remote microphone signal to the CI sound processor. The purpose of this study was to compare the potential benefit of an adaptive multiple-microphone beamformer remote microphone system and a multi-talker network remote microphone system.
Twenty recipients, ages 12 to 84 years, with Advanced Bionics CIs completed sentence-recognition-in-noise tasks while seated at a desk surrounded by three loudspeakers at 0, 90, and 270 degrees. These speakers randomly presented the target speech while competing noise was presented from four loudspeakers located in the corners of the room. Testing was completed in three conditions: 1) CI alone, 2) Remote microphone system with an adaptive multiple-microphone beamformer, and 3) and a multi-talker network remote microphone system each with five different signal levels (15 total conditions).
Significant differences were found across all signal levels and technology conditions. Relative to the CI alone, sentence recognition improvements ranged from 14-23 percentage points with the adaptive multiple-microphone beamformer and 27-47 percentage points with the multi-talker network with superior performance for the latter remote microphone system.
Both remote microphone systems significantly improved speech recognition in noise of CI recipients when listening in small group settings, but the multi-talker network provided superior performance.
人工耳蜗（CI）植入者在有多个说话者的小群体环境中，经常在噪声环境中遇到言语识别困难。在传统的远程麦克风系统中，一个说话者佩戴一个远程麦克风，该麦克风通过无线方式将语音传输到 CI 处理器。该系统将无法传输小群体中多个说话者的信号。然而，具有自适应波束成形功能的多麦克风远程麦克风系统可能对有多个说话者的小群体情况有益。具体来说，一个带有自适应多麦克风波束成形器的远程麦克风可以放置在小群体的中心，波束（即极域）可以自动指向与最有利的语音噪声比相关的方向。然后，远程麦克风的信号可以通过无线方式传输到 CI 声音处理器。或者，小群体中的每个说话者都可以使用一个远程麦克风，该远程麦克风是多说话者网络的一部分，该网络通过无线方式将远程麦克风信号传输到 CI 声音处理器。本研究的目的是比较自适应多麦克风波束成形器远程麦克风系统和多说话者网络远程麦克风系统的潜在优势。
20 名年龄在 12 至 84 岁之间的 Advanced Bionics CI 植入者在一张桌子旁就座，桌子周围有三个扬声器，分别位于 0、90 和 270 度。这些扬声器随机呈现目标语音，同时来自房间四个角落的四个扬声器呈现竞争噪声。测试在三种条件下完成：1）单独使用 CI，2）具有自适应多麦克风波束成形器的远程麦克风系统，以及 3）具有五个不同信号水平的多说话者网络远程麦克风系统（共 15 种条件）。
在所有信号水平和技术条件下均发现显著差异。与单独使用 CI 相比，使用自适应多麦克风波束成形器的句子识别率提高了 14-23 个百分点，使用多说话者网络的句子识别率提高了 27-47 个百分点，后者的远程麦克风系统表现更好。
当在小群体环境中聆听时，两种远程麦克风系统都显著提高了 CI 植入者在噪声中的言语识别能力，但多说话者网络提供了更好的性能。
