《计算机应用研究》|Application Research of Computers

基于多层次注意力机制一维DenseNet音频事件检测

Sound event detection based on 1D DenseNet with multi-level attention

免费全文下载 (已被下载 次)  
获取PDF全文
作者 杨吕祥,胡燕
机构 武汉理工大学 计算机科学与技术学院,武汉 430070
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2020)06-007-1642-05
DOI 10.19734/j.issn.1001-3695.2018.11.0867
摘要 在音频事件检测任务中,目标音频易受背景噪声等因素的干扰,并且其在音频信号流中存在的比例不高,针对这些问题,提出一种多层次注意力机制一维DenseNet(dense convolutional network)音频事件检测模型。使用一维DenseNet模型进行帧级检测能有效地检测音频事件发生的开始和结束时间;在一维DenseNet模型中引入多层次注意力机制,使得不同模块的感知特性随着网络层数的加深而自适应地变化,因此模型可以在不同的网络层次自动选择和关注重要的目标帧而抑制不相关的背景帧。在DCASE 2017任务2的开发数据集上的实验表明,该方法的整体性能较传统的深度学习方法有进一步提高。
关键词 音频事件检测; 深度学习; DenseNet; 多层次注意力机制
基金项目 湖北省自然科学基金重点类资助项目(2017CFA012)
本文URL http://www.arocmag.com/article/01-2020-06-007.html
英文标题 Sound event detection based on 1D DenseNet with multi-level attention
作者英文名 Yang Lyuxiang, Hu Yan
机构英文名 School of Computer Science & Technology,Wuhan University of Technology,Wuhan 430070,China
英文摘要 In sound event detection tasks, the target event was susceptible to background noise, and wasn't present in a significantly high portion of sound signal flow. To solve the problem, this paper proposed a new method of sound event detection based on one-dimensional dense convolutional network(DenseNet) with multi-level attention mechanism. Firstly, it used the one-dimensional DenseNet for frame-level detection, which was effective in finding the precise onset and offset time. Then, it introduced the multi-level attention mechanism in the one-dimensional DenseNet model, which made the attention-aware features from different modules change adaptively as layers went deeper. Therefore, the model could automatically select and attend on important frames for the targets while ignoring the unrelated parts(e. g. the background noise segments). Finally, this paper evaluated the model using DCASE 2017 Task 2 development dataset. Results show that the overall performance of the proposed method has further improvement than the conventional deep learning method.
英文关键词 sound event detection; deep learning; DenseNet; multi-level attention mechanism
参考文献 查看稿件参考文献
 
收稿日期 2018/11/16
修回日期 2019/1/22
页码 1642-1646
中图分类号 TP391.42
文献标志码 A