Musical Instrument Recognition in Polyphonic Audio Through Convolutional Neural Networks and Spectrograms

About this item

Title

Musical Instrument Recognition in Polyphonic Audio Through Convolutional Neural Networks and Spectrograms

Content partner

Auckland University of Technology

Collection

Tuwhera

Description

This study investigates the task of identifying musical instruments in polyphonic compositions using Convolutional Neural Networks (CNNs) from spectrogram inputs, focusing on binary classification. The model showed promising results, with an accuracy of 97% on solo instrument recognition. When applied to polyphonic combinations of 1 to 10 instruments, the overall accuracy was 64%, reflecting the increasing challenge with larger ensembles. These findings contribute to the field of Music Inform...

Read more

Format

Research paper

Research format

Conference item

Date created

2024-07-04

Creator

Rujia, Chen / Ghobakhlou, Ali / Narayanan, Ajit

URL

https://openrepository.aut.ac.nz/handle/10292/17794

Related subjects

binary classifier, CNN

What is the copyright status of this item?

Creative Commons

This item appears to be licensed under a Creative Commons Attribution (BY) 4.0 CC International license.

More Information

Auckland University of Technology has this to say about the rights status of this item:

© 2024 World Academy of Science, Engineering and Technology. Creative Commons Attribution 4.0 International License.

You can learn more about the rights status of this item at: https://creativecommons.org/licenses/by/4.0/

What can I do with this item?

You must always check with Auckland University of Technology to confirm the specific terms of use, but this is our understanding:

Non-infringing use

NZ Copyright law does not prevent every use of a copyright work. You should consider what you can and cannot do with a copyright work.

Share it

This item is suitable for copying and sharing with others, without further permission.

Modify it

This item is suitable for modifying, remixing and building upon, without further permission.

Use it commercially

This item is suitable for commercial use, without further permission.