基于梅尔倒谱系数、深层卷积和Bagging的环境音分类方法

doi:10.11772/j.issn.1001-9081.2019040678

计算机应用 ›› 2019, Vol. 39 ›› Issue (12): 3515-3521.DOI: 10.11772/j.issn.1001-9081.2019040678

基于梅尔倒谱系数、深层卷积和Bagging的环境音分类方法

王天锐, 鲍骞月, 秦品乐

中北大学大数据学院, 太原 030051

收稿日期:2019-04-22 修回日期:2019-07-07 出版日期:2019-12-10 发布日期:2019-12-17
作者简介:王天锐(1997-),男,四川成都人,主要研究方向:深度学习、机器智能;鲍骞月(1998-),男,山西朔州人,主要研究方向:深度学习、机器视觉;秦品乐(1978-),男,山西长治人,副教授,博士,CCF会员,主要研究方向:大数据、机器视觉、三维重建。

Environmental sound classification method based on Mel-frequency cepstral coefficient, deep convolution and Bagging

WANG Tianrui, BAO Qianyue, QIN Pinle

School of Data Science and Technology, North University of China, Taiyuan Shanxi 030051, China

Received:2019-04-22 Revised:2019-07-07 Online:2019-12-10 Published:2019-12-17
Contact: 秦品乐

摘要/Abstract

摘要： 针对传统环境音分类模型对环境音特征提取不充分，以及卷积神经网络用于环境音分类时全连接层易造成过拟合现象的问题，提出了梅尔倒谱系数（MFCC）、深层卷积和Bagging算法相结合的环境音分类方法。首先，针对原始音频文件，利用预加重、加窗、离散傅里叶变换、梅尔滤波器转换、离散余弦映射等方法建立梅尔倒谱系数特征模型；然后，将特征模型输入卷积深度网络进行第二次特征提取；最后，借鉴强化学习思想，用Bagging集成算法集成线性判别分析器、支持向量机（SVM）、Softmax回归、XGBoost四个模型，以投票预测的形式对网络输出结果进行预测。实验结果表明，所提方法能够有效提高对环境音的特征提取能力和深层网络在环境音分类上的抗过拟合能力。

关键词: 环境音分类, 梅尔频率倒谱系数, Bagging集成算法, 特征提取, 深度学习

Abstract: The traditional environmental sound classification model does not fully extract the features of environmental sound, and the full connection layer of conventional neural network is easy to cause over-fitting when the network is used for environmental sound classification. In order to solve the problems, an environmental sound classification method combining with Mel-Frequency Cepstral Coefficient (MFCC), deep convolution and Bagging algorithm was proposed. Firstly, for the original audio file, the MFCC model was established by using pre-emphasis, windowing, discrete Fourier transform, Mel filter transformation, discrete cosine mapping. Secondly, the feature model was input into the convolutional depth network for the second feature extraction. Finally, based on reinforcement learning, the Bagging algorithm was adopted to integrate the linear discriminant analyzer, Support Vector Machine (SVM), softmax regression and eXtreme Gradient Boost (XGBoost) models to predict the network output results by voting prediction. The experimental results show that, the proposed method can effectively improve the feature extraction ability of environmental sound and the anti-over-fitting ability of deep network in environmental sound classification.

Key words: environmental sound classification, Mel-Frequency Cepstral Coefficient (MFCC), Bagging integration algorithm, feature extraction, deep learning

中图分类号:

TP183

王天锐, 鲍骞月, 秦品乐. 基于梅尔倒谱系数、深层卷积和Bagging的环境音分类方法[J]. 计算机应用, 2019, 39(12): 3515-3521.

WANG Tianrui, BAO Qianyue, QIN Pinle. Environmental sound classification method based on Mel-frequency cepstral coefficient, deep convolution and Bagging[J]. Journal of Computer Applications, 2019, 39(12): 3515-3521.

参考文献

[1] PILLOS A, ALGHAMIDI K, ALZAMEL N, et al. A real-time environmental sound recognition system for the Android OS[EB/OL].[2019-02-20]. http://www.cs.tut.fi/sgn/arg/dcase2016/documents/workshop/Pillos-DCASE2016workshop.pdf.
[2] BRIGGS F, RAICH R, FERN X Z. Audio classification of bird species:a statistical manifold approach[C]//Proceedings of the 9th IEEE International Conference on Data Mining. Piscataway:IEEE, 2009:51-60.
[3] LOGAN B. Mel-frequency cepstral coefficients for music modeling[EB/OL].[2019-02-20]. https://www.researchgate.net/publication/2552483_Mel_Frequency_Cepstral_Coefficients_for_Music_Modeling.
[4] CHEN J, KAM A H, ZHANG J, et al. Bathroom activity monitoring based on sound[C]//Proceedings of the 2005 International Conference on Pervasive Computing, LNCS 3468. Berlin:Springer, 2009:47-61.
[5] 李玲俐.基于MFCC-SVM和交叉验证方法的环境音分类[J].计算机与现代化,2016(8):36-39.(LI L L. Environmental sound classification based on MFCC-SVM and cross validation method[J]. Computer and Modernization, 2016(8):36-39.)
[6] 郑继明,魏国华,吴渝.有效的基于内容的音频特征提取方法[J].计算机工程与应用,2009,45(12):131-133,137.(ZHENG J M, WEI G H, WU Y. New effective method on content based audio feature extraction[J]. Computer Engineering and Applications, 2009, 45(12):131-133, 137.)
[7] 付炜,杨洋.基于卷积神经网络和随机森林的音频分类方法[J].计算机应用,2018,38(S2):58-62.(FU W, YANG Y. Audio classification method based on convolutional neural network and random forest[J]. Journal of Computer Applications, 2018, 38(S2):58-62.)
[8] 余清清.噪音环境下基于时-频特征的生态环境声音的分类[J].计算机与数字工程,2017,45(1):8-14,106.(YU Q Q. Eco-environmental sounds classification with time-frequency features under noise conditions[J]. Computer and Digital Engineering, 2017, 45(1):8-14, 106.)
[9] 芮瑞,鲍长春.噪声环境下的窄带音频信号快速分类方法[J].计算机工程与应用,2011,47(16):22-25.(RUI R, BAO C C. Fast classification method of narrow-band audio signals under noisy environment[J]. Computer Engineering and Applications, 2011, 47(16):22-25.)
[10] PICZAK K J. Environmental sound classification with convolutional neural networks[C]//Proceedings of the IEEE 25th International Workshop on Machine Learning for Signal Processing. Piscataway:IEEE, 2015:1-6.
[11] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems. Red Hook:Curran Associates Inc., 2012:1097-1105.
[12] KUMAR A, KHADKEVICH M, FUGEN C. Knowledge transfer from weakly labeled audio using convolutional neural network for sound events and scenes[EB/OL].[2019-03-05]. https://arxiv.org/pdf/1711.01369.pdf.
[13] BREIMAN L. Bagging predictors[J]. Machine Learning, 1996, 24(2):123-40.
[14] 刘亚冲,唐智灵.基于Softmax回归的通信辐射源特征分类识别方法[J].计算机工程,2018,44(2):98-102.(LIU Y C, TANG Z L. Classification and identification method of communication radiation source feature based on softmax regression[J]. Computer Engineering, 2018, 44(2):98-102.)
[15] CHEN T, GUESTRIN C. XGBoost:a scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM, 2016:785-794.
[16] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].[2019-03-05]. https://arxiv.org/pdf/1409.1556.pdf.
[17] IOFFE S, SZEGEDY C. Batch normalization:accelerating deep network training by reducing internal covariate shift[EB/OL].[2019-03-05]. https://arxiv.org/pdf/1502.03167.pdf.
[18] PICZAK K J. ESC:dataset for environmental sound classification[C]//Proceedings of the 23rd ACM International Conference on Multimedia. New York:ACM, 2015:1015-1018.
[19] SAILOR H B, PATIL H A. Filterbank learning using convolutional restricted Boltzmann machine for speech recognition[C]//Proceedings of the 2016 IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway:IEEE, 2016:5895-5899.
[20] TOKOZUME Y, USHIKU Y, HARADA T. Learning from between-class examples for deep sound recognition[EB/OL].[2019-03-05]. https://arxiv.org/pdf/1711.10282.pdf.
[21] TAK R N, AGRAWAL D M, PATIL H A. Novel phase encoded mel filterbank energies for environmental sound classification[C]//Proceedings of the 2017 International Conference on Pattern Recognition and Machine Intelligence, LNCS 10597. Cham:Springer, 2017:317-325.

基于梅尔倒谱系数、深层卷积和Bagging的环境音分类方法

Environmental sound classification method based on Mel-frequency cepstral coefficient, deep convolution and Bagging

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	郑志强, 胡鑫, 翁智, 王雨禾, 程曦. 基于改进DenseNet的牛眼图像特征提取方法[J]. 计算机应用, 2021, 41(9): 2780-2784.
[2]	谢德峰, 吉建民. 融入句法感知表示进行句法增强的语义解析[J]. 计算机应用, 2021, 41(9): 2489-2495.
[3]	代雨柔, 杨庆, 张凤荔, 周帆. 基于自监督学习的社交网络用户轨迹预测模型[J]. 计算机应用, 2021, 41(9): 2545-2551.
[4]	陈成瑞, 孙宁, 何世彪, 廖勇. 面向C-V2X通信的基于深度学习的联合信道估计与均衡算法[J]. 计算机应用, 2021, 41(9): 2687-2693.
[5]	赵宏, 孔东一. 图像特征注意力与自适应注意力融合的图像内容中文描述[J]. 计算机应用, 2021, 41(9): 2496-2503.
[6]	徐江浪, 李林燕, 万新军, 胡伏原. 结合目标检测的室内场景识别方法[J]. 计算机应用, 2021, 41(9): 2720-2725.
[7]	何正海, 线岩团, 王蒙, 余正涛. 融合句法指导与字符注意力机制的案情阅读理解方法[J]. 计算机应用, 2021, 41(8): 2427-2431.
[8]	曹玉红, 徐海, 刘荪傲, 王紫霄, 李宏亮. 基于深度学习的医学影像分割研究综述[J]. 计算机应用, 2021, 41(8): 2273-2287.
[9]	秦斌斌, 彭良康, 卢向明, 钱江波. 司机分心驾驶检测研究进展[J]. 计算机应用, 2021, 41(8): 2330-2337.
[10]	高钦泉, 黄炳城, 刘文哲, 童同. 基于改进CenterNet的竹条表面缺陷检测方法[J]. 计算机应用, 2021, 41(7): 1933-1938.
[11]	侯笑晗, 金国栋, 谭力宁, 薛远亮. 基于自适应和最优特征的合成孔径雷达舰船检测方法[J]. 计算机应用, 2021, 41(7): 2150-2155.
[12]	李亚芳, 梁烨, 冯韦玮, 祖宝开, 康玉健. 基于社区优化的深度网络嵌入方法[J]. 计算机应用, 2021, 41(7): 1956-1963.
[13]	王月, 江逸茗, 兰巨龙. 基于改进三元组网络和K近邻算法的入侵检测[J]. 计算机应用, 2021, 41(7): 1996-2002.
[14]	杜炎, 吕良福, 焦一辰. 基于模糊推理的模糊原型网络[J]. 计算机应用, 2021, 41(7): 1885-1890.
[15]	黄梨, 卢龙. 基于长距离依赖编码与深度残差U-Net的缺血性卒中病灶分割[J]. 计算机应用, 2021, 41(6): 1820-1827.