Performance of machine learning classifiers in distress keywords recognition for audio surveillance applications
The ability to recognize distress speech is the essence of an intelligent audio surveillance system. With this ability, the surveillance system can be configured to detect specific distress keywords and launch appropriate actions to prevent unwanted incidents from progressing. This paper aims to fin...
Saved in:
Main Authors: | , , |
---|---|
Format: | Proceedings |
Language: | English English |
Published: |
Institute of Electrical and Electronics Engineers
2021
|
Subjects: | |
Online Access: | https://eprints.ums.edu.my/id/eprint/32526/1/Performance%20of%20machine%20learning%20classifiers%20in%20distress%20keywords%20recognition%20for%20audio%20surveillance%20applications.ABSTRACT.pdf https://eprints.ums.edu.my/id/eprint/32526/2/Performance%20of%20machine%20learning%20classifiers%20in%20distress%20keywords%20recognition%20for%20audio%20surveillance%20applications.pdf https://eprints.ums.edu.my/id/eprint/32526/ https://ieeexplore.ieee.org/document/9573852 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.ums.eprints.32526 |
---|---|
record_format |
eprints |
spelling |
my.ums.eprints.325262022-05-03T13:31:30Z https://eprints.ums.edu.my/id/eprint/32526/ Performance of machine learning classifiers in distress keywords recognition for audio surveillance applications Nadhirah Johari Mazlina Mamat Ali Chekima Q1-390 Science (General) TK7885-7895 Computer engineering. Computer hardware The ability to recognize distress speech is the essence of an intelligent audio surveillance system. With this ability, the surveillance system can be configured to detect specific distress keywords and launch appropriate actions to prevent unwanted incidents from progressing. This paper aims to find potential distress keywords that the audio surveillance system could recognize. The idea is to use a machine learning classifier as the recognition engine. Five distress keywords: ‘Help’, ‘No’, ‘Oi’, ‘Please’, and ‘Tolong’ were selected to be analyzed. A total of 515 audio signals comprising these five distress keywords were collected and used in the training and testing of 27 classifier models, derived from the Decision Tree, Naïve Bias, Support Vector Machine, K-Nearest Neighbour, Ensemble, and Artificial Neural Network. The features extracted from each audio signal are the Mel-frequency Cepstral Coefficients, while the Principal Component Analysis was applied for feature reduction. The results show that the keyword 'Please' is the most recognized, followed by ‘Help’, ‘Oi’, ‘No’ and ‘Tolong’, respectively. This observation was achieved using the Ensemble Bagged Trees classifier, which can recognize ‘Please’ with 99% accuracy in training and 100% accuracy in testing. Institute of Electrical and Electronics Engineers 2021 Proceedings PeerReviewed text en https://eprints.ums.edu.my/id/eprint/32526/1/Performance%20of%20machine%20learning%20classifiers%20in%20distress%20keywords%20recognition%20for%20audio%20surveillance%20applications.ABSTRACT.pdf text en https://eprints.ums.edu.my/id/eprint/32526/2/Performance%20of%20machine%20learning%20classifiers%20in%20distress%20keywords%20recognition%20for%20audio%20surveillance%20applications.pdf Nadhirah Johari and Mazlina Mamat and Ali Chekima (2021) Performance of machine learning classifiers in distress keywords recognition for audio surveillance applications. https://ieeexplore.ieee.org/document/9573852 |
institution |
Universiti Malaysia Sabah |
building |
UMS Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Malaysia Sabah |
content_source |
UMS Institutional Repository |
url_provider |
http://eprints.ums.edu.my/ |
language |
English English |
topic |
Q1-390 Science (General) TK7885-7895 Computer engineering. Computer hardware |
spellingShingle |
Q1-390 Science (General) TK7885-7895 Computer engineering. Computer hardware Nadhirah Johari Mazlina Mamat Ali Chekima Performance of machine learning classifiers in distress keywords recognition for audio surveillance applications |
description |
The ability to recognize distress speech is the essence of an intelligent audio surveillance system. With this ability, the surveillance system can be configured to detect specific distress keywords and launch appropriate actions to prevent unwanted incidents from progressing. This paper aims to find potential distress keywords that the audio surveillance system could recognize. The idea is to use a machine learning classifier as the recognition engine. Five distress keywords: ‘Help’, ‘No’, ‘Oi’, ‘Please’, and ‘Tolong’ were selected to be analyzed. A total of 515 audio signals comprising these five distress keywords were collected and used in the training and testing of 27 classifier models, derived from the Decision Tree, Naïve Bias, Support Vector Machine, K-Nearest Neighbour, Ensemble, and Artificial Neural Network. The features extracted from each audio signal are the Mel-frequency Cepstral Coefficients, while the Principal Component Analysis was applied for feature reduction. The results show that the keyword 'Please' is the most recognized, followed by ‘Help’, ‘Oi’, ‘No’ and ‘Tolong’, respectively. This observation was achieved using the Ensemble Bagged Trees classifier, which can recognize ‘Please’ with 99% accuracy in training and 100% accuracy in testing. |
format |
Proceedings |
author |
Nadhirah Johari Mazlina Mamat Ali Chekima |
author_facet |
Nadhirah Johari Mazlina Mamat Ali Chekima |
author_sort |
Nadhirah Johari |
title |
Performance of machine learning classifiers in distress keywords recognition for audio surveillance applications |
title_short |
Performance of machine learning classifiers in distress keywords recognition for audio surveillance applications |
title_full |
Performance of machine learning classifiers in distress keywords recognition for audio surveillance applications |
title_fullStr |
Performance of machine learning classifiers in distress keywords recognition for audio surveillance applications |
title_full_unstemmed |
Performance of machine learning classifiers in distress keywords recognition for audio surveillance applications |
title_sort |
performance of machine learning classifiers in distress keywords recognition for audio surveillance applications |
publisher |
Institute of Electrical and Electronics Engineers |
publishDate |
2021 |
url |
https://eprints.ums.edu.my/id/eprint/32526/1/Performance%20of%20machine%20learning%20classifiers%20in%20distress%20keywords%20recognition%20for%20audio%20surveillance%20applications.ABSTRACT.pdf https://eprints.ums.edu.my/id/eprint/32526/2/Performance%20of%20machine%20learning%20classifiers%20in%20distress%20keywords%20recognition%20for%20audio%20surveillance%20applications.pdf https://eprints.ums.edu.my/id/eprint/32526/ https://ieeexplore.ieee.org/document/9573852 |
_version_ |
1760231038410293248 |
score |
13.160551 |