首页> 外文期刊>Pattern recognition letters >Nonnegative features of spectro-temporal sounds for classification
【24h】

Nonnegative features of spectro-temporal sounds for classification

机译:时空声音的非负性特征用于分类

获取原文
获取原文并翻译 | 示例
           

摘要

A parts-based representation is a way of understanding object recognition in the brain. The nonnegative matrix factorization (NMF) is an algorithm which is able to learn a parts-based representation by allowing only non-subtractive combinations [Lee, D.D., Seung, H.S., 1999. Learning the parts of objects by non-negative matrix factorization. Nature 401, 788-791]. In this paper we incorporate a parts-based representation of spectro-temporal sounds into the acoustic feature extraction, which leads to nonnegative features. We present a method of inferring encoding variables in the framework of NMF and show that the method produces robust acoustic features in the presence of noise in the task of general sound classification. Experimental results confirm that the proposed feature extraction method improves the classification performance, especially in the presence of noise, compared to independent component analysis (ICA) which produces holistic features.
机译:基于零件的表示形式是理解大脑中对象识别的一种方式。非负矩阵分解(NMF)是一种算法,该算法能够通过仅允许非减法组合来学习基于零件的表示形式[Lee,D.D.,Seung,H.S.,1999.通过非负矩阵分解来学习对象的各部分。 Nature 401,788-791]。在本文中,我们将基于分时的频谱时声音表示结合到声学特征提取中,这会导致非负性特征。我们提出了一种在NMF框架中推断编码变量的方法,并表明在一般声音分类任务中,该方法在存在噪声的情况下会产生鲁棒的声学特征。实验结果证实,与产生整体特征的独立成分分析(ICA)相比,所提出的特征提取方法可改善分类性能,尤其是在存在噪声的情况下。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号