Saved in:
Bibliographic Details
Main Authors: Dehaghani, Parinaz Binandeh, Pena, Danilo, Aguiar, A. Pedro
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.09802
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914154436100096
author Dehaghani, Parinaz Binandeh
Pena, Danilo
Aguiar, A. Pedro
author_facet Dehaghani, Parinaz Binandeh
Pena, Danilo
Aguiar, A. Pedro
contents This paper explores the impact of dimensionality reduction and pooling methods for Environmental Sound Classification (ESC) using lightweight CNNs. We evaluate Sparse Salient Region Pooling (SSRP) and its variants, SSRP-Basic (SSRP-B) and SSRP-Top-K (SSRP-T), under various hyperparameter settings and compare them with Principal Component Analysis (PCA). Experiments on the ESC-50 dataset demonstrate that SSRP-T achieves up to 80.69 % accuracy, significantly outperforming both the baseline CNN (66.75 %) and the PCA-reduced model (37.60 %). Our findings confirm that a well-tuned sparse pooling strategy provides a robust, efficient, and high-performing solution for ESC tasks, particularly in resource-constrained scenarios where balancing accuracy and computational cost is crucial.
format Preprint
id arxiv_https___arxiv_org_abs_2511_09802
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Investigation of Feature Selection and Pooling Methods for Environmental Sound Classification
Dehaghani, Parinaz Binandeh
Pena, Danilo
Aguiar, A. Pedro
Signal Processing
Sound
This paper explores the impact of dimensionality reduction and pooling methods for Environmental Sound Classification (ESC) using lightweight CNNs. We evaluate Sparse Salient Region Pooling (SSRP) and its variants, SSRP-Basic (SSRP-B) and SSRP-Top-K (SSRP-T), under various hyperparameter settings and compare them with Principal Component Analysis (PCA). Experiments on the ESC-50 dataset demonstrate that SSRP-T achieves up to 80.69 % accuracy, significantly outperforming both the baseline CNN (66.75 %) and the PCA-reduced model (37.60 %). Our findings confirm that a well-tuned sparse pooling strategy provides a robust, efficient, and high-performing solution for ESC tasks, particularly in resource-constrained scenarios where balancing accuracy and computational cost is crucial.
title Investigation of Feature Selection and Pooling Methods for Environmental Sound Classification
topic Signal Processing
Sound
url https://arxiv.org/abs/2511.09802