Adaptive Multi-Scale Detection of Acoustic Events

Abstract

The goal of acoustic (or sound) events detection (AED or SED) is to predict the temporal position of target events in given audio segments. This task plays a significant role in safety monitoring, acoustic early warning and other scenarios. However, the deficiency of data and diversity of acoustic event sources make the AED task a tough issue, especially for prevalent data-driven methods. In this paper, we start from analyzing acoustic events according to their time-frequency domain properties, showing that different acoustic events have different time-frequency scale characteristics. Inspired by the analysis, we propose an adaptive multi-scale detection (AdaMD) method. By taking advantage of hourglass neural network and gated recurrent unit (GRU) module, our AdaMD produces multiple predictions at different temporal and frequency resolutions. An adaptive training algorithm is subsequently adopted to combine multi-scale predictions to enhance the overall capability. Experimental results on Detection and Classification of Acoustic Scenes and Events 2017 (DCASE 2017) Task 2, DCASE 2016 Task 3 and DCASE 2017 Task 3 demonstrate that the AdaMD outperforms published state-of-the-art competitors in terms of the metrics of event error rate (ER) and F1-score. The verification experiment on our collected factory mechanical dataset also proves the noiseresistant capability of the AdaMD, providing the possibility for it to be deployed in the complex environment. Index Terms-rare acoustic event detection, adaptive multiscale, hourglass network PLACE PHOTO HERE Liang He Liang HE received the B.S. degree in communication engineering from Civil Aviation

Visual Object Detector for Cow Sound Event Detection

“…This section details the architectural details and evaluation metrics for rare SED using CNN and visual object detectors. A simple CNN architecture was selected in this experiment that did not involve the use of any CNN variant such as guided learning [2], a recurrent neural network [5,6], or weakly-supervised learning [21,22]. These architectures have provided best results in previous challenges on preexisting data and included annotations.…”

“…Among these, the recurrent neural network (RNN)-based SED method [3,4] adopts a large temporal context and performs relatively better than basic CNN structure. Hybrid structures containing both CNN and RNN layers, known as convolutional recurrent neural networks (CRNNs) [5,6] are being developed for SED that integrate both spatial and C temporal properties of the audio signal. A CRNN for joint sound event localization and detection (SELD) of multiple overlapping sound events in three dimensional (3D) space has been proposed by Adavanne et al [7].…”

A Comprehensive Review of Polyphonic Sound Event Detection

“…Ding and He [95] proposed an adaptive multi-scale detection method that combined the idea of an hourglass network with Bidirectional GRU (BGR). It is a CRNN but with much higher sophistication.…”

“…The resulting values were then added up and sent to the output layer. In their study, Ding and He [95] proposed using a 4 layers hourglass network with 3 layers bidirectional GRU at each scale. Based on such architecture, Ding and He [95] achieved a single second F1-score of 48.7% with an ER of 0.7821 on TUT-SED 2016 evaluation dataset and a single second F1-score of 43.6% with an ER of 0.7723 on TUT-SED 2017 evaluation dataset.…”

Fitting Analysis of Inland Ship Fuel Consumption Considering Navigation Status and Environmental Factors

“…The RNNs have been successfully applied to various tasks, such as data-driven modelling [39], image caption [40], sentiment analysis [41] and speech recognition [42]. For example, Ergen and Kozat [43] studied online training of the LSTM architecture in a distributed network of nodes for regression and introduced online distributed training algorithms for variable-length data sequences. Zhao et al [44] proposed a new approach, the CAM-RNN to extract the most correlated visual feature and a text feature for the task of video captioning, which was composed of three parts, i.e., visual attention module, text attention module and balancing gate.…”