Histogram of Gradients of Time–Frequency Representations for Audio Scene Classification
Histogram of Gradients of Time–Frequency Representations for Audio Scene Classification
复制标题
DOI：
10.1109/taslp.2014.2375575
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
A. Rakotomamonjy;G. Gasso
中科院分区：
文献类型：
--
作者：
A. Rakotomamonjy;G. Gasso
介绍了我们的进入检测和分类的声学场景的挑战。我们提出的声学场景分类方法是基于将音频信号转换成时频表示，然后提取有关形状和时频结构演变的相关特征。这些特征基于梯度直方图，随后将其馈送到多类线性支持向量机。
Presents our entry to the Detection and Classification of Acoustic Scenes challenge. The approach we propose for classifying acoustic scenes is based on transforming the audio signal into a time-frequency representation and then in extracting relevant features about shapes and evolutions of time-frequency structures. These features are based on histogram of gradients that are subsequently fed to a multi-class linear support vector machines.