Coordination Funds

在日常生活中，我们被大量的声音事件所包围，比如人们说话，音乐的播放，以及各种各样的噪音。然而，我们能够毫不费力地在这样的环境中交谈，在不考虑其他声音的情况下找回想要的声音，或者根据观察到的声音场景，得出关于环境和其中活动的组成的结论。具有类似功能的技术系统将在智能家居、环境辅助生活、个人通信和监控等领域得到广泛应用。随着声传感器成本的不断降低以及无线网络和移动设备的普及，强大的无线声传感器网络的技术基础设施是可用的，而释放新的颠覆性应用的瓶颈显然是在算法方面。本研究单元研究声学信号处理和声学传感器网络分类的解决方案和限制。我们的目标是解决目前的缺点，并开发一个通用平台，使自动神经网络更能适应声环境和传感器配置的变化，减少对监督的依赖，同时对用户更值得信赖。这将为将先进的声学信号处理与音频语义分析相结合的新一类应用铺平道路。项目目标将采用三层方法来实现，底层处理通信和同步方面，中间层处理信号提取和增强方面，上层处理保护隐私的声学场景分类和解释方面。为了完成这些任务，我们将融合先进的统计信号处理和机器学习技术。我们的项目将开创一个通用的、通用的无线声学传感器网络框架。这个框架支持几种声学应用，包括最先进的和新兴的。

In daily life, we are surrounded by a multitude of acoustic events, such as people speaking, the playback of music, and noises of all kinds. We are nevertheless able to effortlessly converse in such an environment, retrieve a desired voice while disregarding others, or draw conclusions about the composition of the environment and activities therein, given the observed sound scene. A technical system with similar capabilities would find numerous applications in fields as diverse as smart homes, ambient-assisted living, personal communications, and surveillance. With the continuously decreasing cost of acoustic sensors and the pervasiveness of wireless networks and mobile devices, the technological infrastructure for powerful wireless acoustic sensor networks is available, and the bottleneck for unleashing new disruptive applications is clearly on the algorithmic side.This Research Unit investigates solutions and limitations for acoustic signal processing and classification over acoustic sensor networks. We aim to tackle current shortcomings and to develop a common plattform which will make ASNs more adaptive to the variability of acoustic environments and sensor configurations, less dependent on supervision, and at the same time more trustworthy for the users. This will pave the way for a new class of applications that combine advanced acoustic signal processing with semantic analysis of audio. The project objectives will be achieved by adopting a three-layer approach treating communication and synchronization aspects on the lower layer, signal extraction and enhancement on the middle layer, and privacy-preserving acoustic scene classification and interpretation on the upper layer. To carry out those tasks we will blend advanced statistical signal processing with machine learning techniques. Our project will pioneer a generic, versatile framework for wireless acoustic sensor networks. This framework supports several classes of acoustic applications, both state-of-the-art and emerging ones.