Citation

Lim, Kian Ming and Lee, Chin Poo and Lee, Zhi Yang and Alqahtani, Ali (2023) EnViTSA: Ensemble of Vision Transformer with SpecAugment for Acoustic Event Classification. Sensors, 23 (22). p. 9084. ISSN 1424-8220

Text 19.pdf - Published Version Restricted to Repository staff only Download (12MB)

Abstract

: Recent successes in deep learning have inspired researchers to apply deep neural networks to Acoustic Event Classification (AEC). While deep learning methods can train effective AEC models, they are susceptible to overfitting due to the models’ high complexity. In this paper, we introduce EnViTSA, an innovative approach that tackles key challenges in AEC. EnViTSA combines an ensemble of Vision Transformers with SpecAugment, a novel data augmentation technique, to significantly enhance AEC performance. Raw acoustic signals are transformed into Log Mel-spectrograms using Short-Time Fourier Transform, resulting in a fixed-size spectrogram representation. To address data scarcity and overfitting issues, we employ SpecAugment to generate additional training samples through time masking and frequency masking. The core of EnViTSA resides in its ensemble of pretrained Vision Transformers, harnessing the unique strengths of the Vision Transformer architecture. This ensemble approach not only reduces inductive biases but also effectively mitigates overfitting. In this study, we evaluate the EnViTSA method on three benchmark datasets: ESC-10, ESC-50, and UrbanSound8K. The experimental results underscore the efficacy of our approach, achieving impressive accuracy scores of 93.50%, 85.85%, and 83.20% on ESC-10, ESC-50, and UrbanSound8K, respectively. EnViTSA represents a substantial advancement in AEC, demonstrating the potential of Vision Transformers and SpecAugment in the acoustic domain.

Item Type:	Article
Uncontrolled Keywords:	Deep learning, neural network
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	02 Jan 2024 08:57
Last Modified:	02 Jan 2024 08:57
URII:	http://shdl.mmu.edu.my/id/eprint/11977

Downloads

Edit (login required)