Go to:
Logótipo
Você está em: Start » Publications » View » Transformers for Urban Sound Classification-A Comprehensive Performance Evaluation
Publication

Transformers for Urban Sound Classification-A Comprehensive Performance Evaluation

Title
Transformers for Urban Sound Classification-A Comprehensive Performance Evaluation
Type
Article in International Scientific Journal
Year
2022-11
Authors
Ana Filipa Rodrigues Nogueira
(Author)
Other
View Personal Page You do not have permissions to view the institutional email. Search for Participant Publications Without AUTHENTICUS Without ORCID
Hugo S. Oliveira
(Author)
Other
View Personal Page You do not have permissions to view the institutional email. Search for Participant Publications Without AUTHENTICUS Without ORCID
José J. M. Machado
(Author)
FEUP
View Personal Page You do not have permissions to view the institutional email. Search for Participant Publications View Authenticus page View ORCID page
João Manuel R. S. Tavares
(Author)
FEUP
View Personal Page You do not have permissions to view the institutional email. Search for Participant Publications View Authenticus page View ORCID page
Journal
Title: SensorsImported from Authenticus Search for Journal Publications
Vol. 22 No. 4
Pages: 8874-8874
ISSN: 1424-3210
Publisher: MDPI
Indexing
Publicação em ISI Web of Knowledge ISI Web of Knowledge - 0 Citations
Publicação em ISI Web of Science ISI Web of Science
Clarivate Analytics
Other information
Authenticus ID: P-00X-JV1
Abstract (EN): Many relevant sound events occur in urban scenarios, and robust classification models are required to identify abnormal and relevant events correctly. These models need to identify such events within valuable time, being effective and prompt. It is also essential to determine for how much time these events prevail. This article presents an extensive analysis developed to identify the best-performing model to successfully classify a broad set of sound events occurring in urban scenarios. Analysis and modelling of Transformer models were performed using available public datasets with different sets of sound classes. The Transformer models' performance was compared to the one achieved by the baseline model and end-to-end convolutional models. Furthermore, the benefits of using pre-training from image and sound domains and data augmentation techniques were identified. Additionally, complementary methods that have been used to improve the models' performance and good practices to obtain robust sound classification models were investigated. After an extensive evaluation, it was found that the most promising results were obtained by employing a Transformer model using a novel Adam optimizer with weight decay and transfer learning from the audio domain by reusing the weights from AudioSet, which led to an accuracy score of 89.8% for the UrbanSound8K dataset, 95.8% for the ESC-50 dataset, and 99% for the ESC-10 dataset, respectively.
Language: English
Type (Professor's evaluation): Scientific
No. of pages: 17
Documents
File name Description Size
paper 1st Page 181.74 KB
sensors-22-08874 Paper 2063.85 KB
Related Publications

Of the same authors

Sound Classification and Processing of Urban Environments: A Systematic Literature Review (2022)
Another Publication in an International Scientific Journal
Ana Filipa Rodrigues Nogueira; Hugo S. Oliveira; José J. M. Machado; João Manuel R. S. Tavares

Of the same journal

yy Optical Fiber Temperature Sensors and Their Biomedical Applications (2020)
Another Publication in an International Scientific Journal
Roriz, P; Susana Silva; Frazao, O; Novais, S
Wearable Health Devices-Vital Sign Monitoring, Systems and Technologies (2018)
Another Publication in an International Scientific Journal
Dias, D; Cunha, JPS
Visualization of Urban Mobility Data from Intelligent Transportation Systems (2019)
Another Publication in an International Scientific Journal
Sobral, T; Teresa Galvão Dias; José Luís Moura Borges
Visual Sensor Networks and Related Applications (2019)
Another Publication in an International Scientific Journal
Costa, DG; Francisco Vasques; Collotta, M
Urban Safety: An Image-Processing and Deep-Learning-Based Intelligent Traffic Management and Control System (2021)
Another Publication in an International Scientific Journal
Selim Reza; Hugo S. Oliveira; José J. M. Machado; João Manuel R. S. Tavares

See all (197)

Recommend this page Top
Copyright 1996-2024 © Faculdade de Medicina da Universidade do Porto  I Terms and Conditions  I Acessibility  I Index A-Z  I Guest Book
Page created on: 2024-08-27 at 14:54:21
Acceptable Use Policy | Data Protection Policy | Complaint Portal | Política de Captação e Difusão da Imagem Pessoal em Suporte Digital