Robust Sound Source Tracking Using SRP-PHAT and 3D Convolutional Neural Networks
University of Zaragoza（萨拉戈萨大学）
David Diaz-Guerra, Antonio Miguel, Jose R. Beltran
备注：
This is a pre-print of an article published in IEEE/ACM Transactions on Audio Speech and Language Processing. The code to reproduce this work can be found in our GitHub repository: https://github.com/DavidDiazGuerra/Cross3D
发表信息：
in IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 29, pp. 300-311, 2021
In this paper, we present a new single sound source DOA estimation and tracking system based on the well-known SRP-PHAT algorithm and a three-dimensional Convolutional Neural Network. It uses SRP-PHAT power maps as input features of a fully convolutional causal architecture that uses 3D convolutional layers to accurately perform the tracking of a sound source even in highly reverberant scenarios where most of the state of the art techniques fail. Unlike previous methods, since we do not use bidirectional recurrent layers and all our convolutional layers are causal in the time dimension, our system is feasible for real-time applications and it provides a new DOA estimation for each new SRP-PHAT map. To train the model, we introduce a new procedure to simulate random trajectories as they are needed during the training, equivalent to an infinite-size dataset with high flexibility to modify its acoustical conditions such as the reverberation time. We use both acoustical simulations on a large range of reverberation times and the actual recordings of the LOCATA dataset to prove the robustness of our system and its good performance even using low-resolution SRP-PHAT maps.
Deep Residual Network for Sound Source Localization in the Time Domain
Skolkovo Institute of Science and Technology（斯科尔科沃科学技术研究院）；Tsinghua University（清华大学）
本研究提出一种使用深度残差神经网络的时域声源定位系统。来自具有3cm间距的线性8通道麦克风阵列的数据由网络用于方向估计。我们建议使用深度残差网络进行声源定位，将定位任务视为分类任务。本研究描述了收集的数据集和神经网络的开发架构。本研究将展示训练过程及其结果。开发的系统进行了测试，验证部分的数据集和新的数据捕获在实时的。对30 m s的声音帧的分类准确率为99.2%。声源定位的标准差为4°。所提出的声源定位方法在语音识别流水线内部进行了测试。与使用GCC-PHAT声源定位的类似语音识别管道相比，其使用降低了1.14%的单词出错率。
Exploiting spatial diversity for increasing the robustness of sound source localization systems against reverberation
Centro de Investigación en Tecnologías Software y Sistemas Multimedia para la Sostenibilidad (CITSEM)（可持续软件技术与多媒体系统研究中心）；Universidad Politécnica de Madrid（马德里理工大学）
A Review on Sound Source Localization in Robotics: Focusing on Deep Learning Methods
Department of Management, Economics and Industrial Engineering, Politecnico di Milano, Via Lambruschini 24/b, 20156, Milan, Italy（管理、经济与工业工程系，米兰理工学院）；Department of Materials and Mechanical Engineering, University of Turku, Vesilinnantie 5, Turku, 20014, Finland（材料与机械工程系，图尔库大学）；Department of Electrical and Computer Engineering, University of Victoria, Victoria, BC V8P 5C2, Canada（电气与计算机工程系，维多利亚大学）
