An Efficient Visual-Based Method for Classifying Instrumental Audio using Deep Learning

Justin Hall, Wesley O'Quinn, Rami J. Haddad

Research output: Contribution to book or proceedingConference articlepeer-review

3 Scopus citations

Abstract

In this paper, an efficient method for classifying and identifying instrumental audio is proposed via utilizing a deep learning image classification algorithm. The method of classification will involve analyzing the visual equivalent of an audio sample with a neural network to identify the generating musical instrument. Audio samples are converted into a logarithmic spectrogram format, which allows visual classifiers to attempt the identification of the audio source. The primary focus is on developing an efficient method for analyzing audio spectrograms using various forms of neural networks and analysis techniques. The use of deep learning convolutional neural networks in analyzing visually formatted audio data provides an enhanced classification method over traditional schemes. A classification accuracy of 73.7% was achieved with a limited data set and minimal manipulation of network architecture.

Original languageEnglish
Title of host publication2019 IEEE SoutheastCon, SoutheastCon 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728101378
DOIs
StatePublished - Apr 2019
Event2019 IEEE SoutheastCon, SoutheastCon 2019 - Huntsville, United States
Duration: Apr 11 2019Apr 14 2019

Publication series

NameConference Proceedings - IEEE SOUTHEASTCON
Volume2019-April
ISSN (Print)1091-0050
ISSN (Electronic)1558-058X

Conference

Conference2019 IEEE SoutheastCon, SoutheastCon 2019
Country/TerritoryUnited States
CityHuntsville
Period04/11/1904/14/19

Scopus Subject Areas

  • Computer Networks and Communications
  • Software
  • Electrical and Electronic Engineering
  • Control and Systems Engineering
  • Signal Processing

Keywords

  • Audio Classification
  • Audio Visual Transform
  • Deep Learning
  • Music Instrument
  • Neural Networks
  • Spectrograms
  • Transfer Learning

Fingerprint

Dive into the research topics of 'An Efficient Visual-Based Method for Classifying Instrumental Audio using Deep Learning'. Together they form a unique fingerprint.

Cite this