Cover Image

Rethinking environmental sound classification using convolutional neural networks: optimized parameter tuning of single feature extraction

The classification of environmental sounds is important for emerging applications such as automatic audio surveillance, audio forensics, and robot navigation. Existing techniques combined multiple features and stacked many CNN layers (very deep learning) to reach the desired accuracy. Instead of usi...

Full description

Saved in:
Bibliographic Details
Main Authors: Al-Hattab, Yousef Abd, Mohd Zaki, Hasan Firdaus, Shafie, Amir Akramin
Format: Article
Language:English
English
Published: Springer Nature 2021
Subjects:
TK7885 Computer engineering
Online Access:http://irep.iium.edu.my/90215/7/90215_Rethinking%20environmental%20sound%20classification%20using%20convolutional%20neural%20networks_SCOPUS.pdf
http://irep.iium.edu.my/90215/8/90215_Rethinking%20environmental%20sound%20classification%20using%20convolutional%20neural%20networks.pdf
http://irep.iium.edu.my/90215/
https://link.springer.com/article/10.1007/s00521-021-06091-7
https://doi.org/10.1007/s00521-021-06091-7
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.iium.irep.90215
record_format dspace
spelling my.iium.irep.902152021-06-15T06:30:27Z http://irep.iium.edu.my/90215/ Rethinking environmental sound classification using convolutional neural networks: optimized parameter tuning of single feature extraction Al-Hattab, Yousef Abd Mohd Zaki, Hasan Firdaus Shafie, Amir Akramin TK7885 Computer engineering The classification of environmental sounds is important for emerging applications such as automatic audio surveillance, audio forensics, and robot navigation. Existing techniques combined multiple features and stacked many CNN layers (very deep learning) to reach the desired accuracy. Instead of using many features and going deeper by stacking layers that are resource extensive, this paper proposes a novel technique that uses only a single feature, namely the Mel-Frequency Cepstral Coefficient (MFCC) and just three layers of CNN. We demonstrate that such a simple network can considerably outperform several conventional and deep learning-based algorithms. Through a carefully and empirically parameters fine-tuning of the data input, we reported a model that is significantly less complex in the architecture yet has recorded a similar accuracy of 95.59% compared to state-of-the-art deep models on UrbanSound8k dataset. We conjecture that our accurate lightweight model is an excellent environmental sound recognizer for the application on resource-constraint embedded platform. Springer Nature 2021-05-26 Article PeerReviewed application/pdf en http://irep.iium.edu.my/90215/7/90215_Rethinking%20environmental%20sound%20classification%20using%20convolutional%20neural%20networks_SCOPUS.pdf application/pdf en http://irep.iium.edu.my/90215/8/90215_Rethinking%20environmental%20sound%20classification%20using%20convolutional%20neural%20networks.pdf Al-Hattab, Yousef Abd and Mohd Zaki, Hasan Firdaus and Shafie, Amir Akramin (2021) Rethinking environmental sound classification using convolutional neural networks: optimized parameter tuning of single feature extraction. Neural Computing and Applications. pp. 1-19. ISSN 0941-0643 E-ISSN 1433-3058 https://link.springer.com/article/10.1007/s00521-021-06091-7 https://doi.org/10.1007/s00521-021-06091-7
institution Universiti Islam Antarabangsa Malaysia
building IIUM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider International Islamic University Malaysia
content_source IIUM Repository (IREP)
url_provider http://irep.iium.edu.my/
language English
English
topic TK7885 Computer engineering
spellingShingle TK7885 Computer engineering
Al-Hattab, Yousef Abd
Mohd Zaki, Hasan Firdaus
Shafie, Amir Akramin
Rethinking environmental sound classification using convolutional neural networks: optimized parameter tuning of single feature extraction
description The classification of environmental sounds is important for emerging applications such as automatic audio surveillance, audio forensics, and robot navigation. Existing techniques combined multiple features and stacked many CNN layers (very deep learning) to reach the desired accuracy. Instead of using many features and going deeper by stacking layers that are resource extensive, this paper proposes a novel technique that uses only a single feature, namely the Mel-Frequency Cepstral Coefficient (MFCC) and just three layers of CNN. We demonstrate that such a simple network can considerably outperform several conventional and deep learning-based algorithms. Through a carefully and empirically parameters fine-tuning of the data input, we reported a model that is significantly less complex in the architecture yet has recorded a similar accuracy of 95.59% compared to state-of-the-art deep models on UrbanSound8k dataset. We conjecture that our accurate lightweight model is an excellent environmental sound recognizer for the application on resource-constraint embedded platform.
format Article
author Al-Hattab, Yousef Abd
Mohd Zaki, Hasan Firdaus
Shafie, Amir Akramin
author_facet Al-Hattab, Yousef Abd
Mohd Zaki, Hasan Firdaus
Shafie, Amir Akramin
author_sort Al-Hattab, Yousef Abd
title Rethinking environmental sound classification using convolutional neural networks: optimized parameter tuning of single feature extraction
title_short Rethinking environmental sound classification using convolutional neural networks: optimized parameter tuning of single feature extraction
title_full Rethinking environmental sound classification using convolutional neural networks: optimized parameter tuning of single feature extraction
title_fullStr Rethinking environmental sound classification using convolutional neural networks: optimized parameter tuning of single feature extraction
title_full_unstemmed Rethinking environmental sound classification using convolutional neural networks: optimized parameter tuning of single feature extraction
title_sort rethinking environmental sound classification using convolutional neural networks: optimized parameter tuning of single feature extraction
publisher Springer Nature
publishDate 2021
url http://irep.iium.edu.my/90215/7/90215_Rethinking%20environmental%20sound%20classification%20using%20convolutional%20neural%20networks_SCOPUS.pdf
http://irep.iium.edu.my/90215/8/90215_Rethinking%20environmental%20sound%20classification%20using%20convolutional%20neural%20networks.pdf
http://irep.iium.edu.my/90215/
https://link.springer.com/article/10.1007/s00521-021-06091-7
https://doi.org/10.1007/s00521-021-06091-7
_version_ 1703960212662124544
score 13.211869

Similar Items