CNNbased musical instrument identification using time-frequency localized features

被引：7

作者：

Dutta, Arindam ^{[1
]}

Sil, Dibakar ^{[2
]}

Chandra, Aniruddha ^{[2
]}

Palit, Sarbani ^{[3
]}

机构：

[1] Natl Inst Technol, EE Dept, Durgapur, India

[2] Natl Inst Technol, ECE Dept, Durgapur 713209, India

[3] Indian Stat Inst, CVPR Unit, Kolkata, India

来源：

INTERNET TECHNOLOGY LETTERS | 2022年 / 5卷 / 01期

关键词：

audio signal classification; convolutional neural network; instrument recognition; scalogram; SEPARATION;

D O I：

10.1002/itl2.191

中图分类号：

TN [电子技术、通信技术];

学科分类号：

0809 ;

摘要：

In this paper, the authors make an attempt to solve the convoluted problem of identifying musical instruments based on their audio excerpts, using a deep convolutional neural network. Continuous wavelet transform of audio signals are realized through Morse wavelet and two-dimensional feature maps are formed, which are then fed to a simple yet robust convolutional neural network. The outcome is appreciable in the sense that training the model with just 20% of the data and testing on the rest gives a classification accuracy of 85%.

引用

页数：6

共 19 条

[1] Automatic musical instrument classification using fractional fourier transform based- MFCC features and counter propagation neural network [J].

Bhalke, D. G. ;

Rao, C. B. Rama ;

Bormane, D. S. .

JOURNAL OF INTELLIGENT INFORMATION SYSTEMS, 2016, 46 (03) :425-446

[2] Musical Source Separation An introduction [J].

Cano, Estefania ;

FitzGerald, Derry ;

Liutkus, Antoine ;

Plumbley, Mark D. ;

Stoeter, Fabian-Robert .

IEEE SIGNAL PROCESSING MAGAZINE, 2019, 36 (01) :31-40

[3]

Chon SH, 2017, EMPIR MUSICOL REV, V12, P116

[4]

Ghosh A, 2018, 2018 3RD INTERNATIONAL CONFERENCE ON ELECTRICAL, ELECTRONICS, COMMUNICATION, COMPUTER, AND OPTIMIZATION TECHNIQUES (ICEECCOT - 2018), P509, DOI 10.1109/ICEECCOT43722.2018.9001486

[5] Deep Convolutional Neural Networks for Predominant Instrument Recognition in Polyphonic Music [J].