Optimal Scalogram for Computational Complexity Reduction in Acoustic Recognition Using Deep Learning
Dang Thoai Phan, Tuan Anh Huynh, Van Tuan Pham, Cao Minh Tran, Van Thuan Mai, Ngoc Quy Tran
TL;DR
This work tackles the high computational cost of Continuous Wavelet Transform (CWT) in acoustic recognition by introducing optCWT, which reduces complexity through adjustable Morlet wavelet length (WL) and output hop size (H) with a grid search to balance accuracy and efficiency. The approach preserves multiresolution benefits while achieving substantial speedups, demonstrated on the MIMII industrial sounds dataset where per-file time drops from 8.09 s to 1.15 s and total dataset time from 122.5 h to 17.5 h, with largely comparable predictive performance to the baseline. The study also analyzes the resulting scalograms, noting sparser energy distributions under optCWT, and discusses limitations and future extensions, including testing other wavelets and broader datasets. Overall, optCWT offers a practical path to real-time, scalable CWT-based acoustic recognition, with clear avenues for generalization and further optimization.
Abstract
The Continuous Wavelet Transform (CWT) is an effective tool for feature extraction in acoustic recognition using Convolutional Neural Networks (CNNs), particularly when applied to non-stationary audio. However, its high computational cost poses a significant challenge, often leading researchers to prefer alternative methods such as the Short-Time Fourier Transform (STFT). To address this issue, this paper proposes a method to reduce the computational complexity of CWT by optimizing the length of the wavelet kernel and the hop size of the output scalogram. Experimental results demonstrate that the proposed approach significantly reduces computational cost while maintaining the robust performance of the trained model in acoustic recognition tasks.