Active sound field control for sound reproduction in open ear canals

开放式听力系统已被证明是轻度至中度听力损失康复的有效解决方案。与封闭拟合系统相比，它们特别容易被最终用户接受，最显著的原因是它们避免了所谓的遮挡效应。另一方面，开放式听力系统在技术上具有挑战性，因为有效开放的耳道使它们对来自外部的噪音渗透和从接收器到麦克风的声音反馈敏感。当前（被动）平衡遮挡、外部噪声和反馈抑制的最佳方法是仔细选择合适的排气孔大小，并结合反馈抵消算法。另外，即将在商业助听器中使用的耳道麦克风使得采用主动方法来解决这些不同的问题成为可能，此外还可以在耳鼓处获得最佳的个性化声压均衡。该项目的长期目标是优化利用耳道或通气孔中麦克风和接收器的可用性，以开发改进的个性化方法，用于鼓膜处的声压均衡，反馈消除和开放式听力设备的降噪。在最初的资助阶段，重点是针对配备一个接收器和一个耳道麦克风的助听器进行个性化均衡、主动降噪和个性化反馈消除，其中这些问题是相互独立处理的，我们考虑了内部声场的时不变情况。在第二个资助期，我们的目标是将开发的方法扩展到1)耳道几何形状变化引起的内部声场的动态场景，以及2)在耳道或通风孔中配备多个接收器和麦克风的助听器，从而实现目标反馈消除和平衡鼓膜处的声压。这包括在初始资助期间开发的个性化耳道和反馈路径模型的适应，以及新型多扬声器-多麦克风信号处理算法和用于组合反馈抵消和均衡的声学模型的开发。此外，还将开发动态变化反馈路径和耳道阻抗的测量和评估程序以及个性化信号处理算法的实时演示。因此，该项目有助于解决研究单位解决的所有主要问题，即使用局部方法（反馈抵消和均衡算法）的呈现问题，个性化问题（个性化耳道和反馈路径模型）以及降噪问题。

Open-fitting hearing systems have proven to be efficient solutions for the rehabilitation of mild to moderate hearing loss. In comparison to closed-fitting systems, they are in particular much better accepted by the end user, most notably because they avoid the so-called occlusion effect. On the other hand, open-fitting hearing systems are technically challenging, because the effectively open ear canal makes them sensitive to noise penetrating from the Outside and to acoustic feedback from the receiver to the microphones. The current (passive) approach to optimally balance the suppression of occlusion, external noise and feedback is to carefully choose an appropriate vent size in combination with feedback cancellation algorithms. Alternatively, the upcoming availability of ear canal microphones in commercial hearing aids makes it possible to employ active approaches to tackle these different problems, and in addition derive an optimal individualized equalization of the sound pressure at the ear drum.The long-term objective of this project is to optimally exploit the availability of microphones and receivers in the ear canal or in the vent to develop improved individualized approaches for sound pressure equalization at the ear drum, for feedback cancellation and for noise reduction in open-fitting hearing devices.In the initial funding period, the focus was on individualized equalization, active noise reduction and individualized feedback cancellation for a hearing aid device equipped with one receiver and one ear canal microphone, where these problems were treated independently of each other and we considered time-invariant scenarios for the internal sound field. In the second funding period, we aim to extend the developed approaches to 1) dynamic scenarios for the internal sound field caused by changes ofthe ear canal geometry, and 2) hearing aid devices equipped with multiple receivers and microphones in the ear canal or in the vent, enabling to target feedback cancellation and equalization of the sound pressure at the ear drum simultaneously. This comprises both an adaptation of the individualized ear canal and feedback path models developed during the initial funding period, as well as the development of novel multi-loudspeaker-multi-microphone signal processing algorithms and acoustic models for combined feedback cancellation and equalization. In addition, measurement and evaluation procedures for dynamically varying feedback paths and acoustic ear canal impedances and real-time demonstrators of the individualized signal processing algorithms will be developed.Hence, this project contributes to all cardinal problems addressed by the Research Unit, namely the presentation problem using local approaches (feedback cancellation and equalization algorithms), the individualization problem (individualized ear canal and feedback path models), and the noise reduction problem.