QQ客服 官方微博 反馈留言

	Robust Sound Event Classification by Using Denoising Autoencoder
	Zhou, Jianchao ; Peng, Liqun ; Chen, Xiaoou ; Yang, Deshun
	2016
英文摘要	Over the last decade, a lot of research has been done on sound event classification. But a main problem with sound event classification is that the performance sharply degrades in the presence of noise. As spectrogram-based image features and denoising autoencoder reportedly have superior performance in noisy conditions, this paper proposes a new robust feature called denoising autoencoder image feature (DIF) for sound event classification which is an image feature extracted from an image-like representation produced by denoising autoencoder. Performance of the feature is evaluated by a classification experiment using a SVM classifier on audio examples with different noise levels, and compared with that of baseline features including mel-frequency cepstral coefficients (MFCC) and spectrogram image feature. The proposed DIF demonstrates better performance under noise-corrupted conditions.; National Hi-Tech Research and Development Program (863 Program) of China [2014AA015102]; Natural Science Foundation of China [61370116]; CPCI-S(ISTP); zhoujc@pku.edu.cn; pengliqun@pku.edu.cn; chenxiaoou@pku.edu.cn; yangdeshun@pku.edu.cn
语种	英语
出处	18th IEEE International Workshop on Multimedia Signal Processing (MMSP)
内容类型	其他
源URL	[http://ir.pku.edu.cn/handle/20.500.11897/459663]
专题	信息科学技术学院
推荐引用方式 GB/T 7714	Zhou, Jianchao,Peng, Liqun,Chen, Xiaoou,et al. Robust Sound Event Classification by Using Denoising Autoencoder. 2016-01-01.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。