Low-complexity CNNs for Acoustic Scene Classification

TL;DR

This paper introduces a low-complexity CNN framework for acoustic scene classification that reduces resource requirements through pruning, quantization, and ensemble methods, achieving competitive performance on a standard dataset.

Contribution

It presents a novel low-complexity CNN architecture with pruning and quantization, combined in an ensemble to enhance acoustic scene classification performance.

Findings

Approximately 60K parameters in the ensemble model

Requires 19 million multiply-accumulate operations

Improves accuracy by 2-4 percentage points over baseline

Abstract

This paper presents a low-complexity framework for acoustic scene classification (ASC). Most of the frameworks designed for ASC use convolutional neural networks (CNNs) due to their learning ability and improved performance compared to hand-engineered features. However, CNNs are resource hungry due to their large size and high computational complexity. Therefore, CNNs are difficult to deploy on resource constrained devices. This paper addresses the problem of reducing the computational complexity and memory requirement in CNNs. We propose a low-complexity CNN architecture, and apply pruning and quantization to further reduce the parameters and memory. We then propose an ensemble framework that combines various low-complexity CNNs to improve the overall performance. An experimental evaluation of the proposed framework is performed on the publicly available DCASE 2022 Task 1 that focuses on ASC. The proposed ensemble framework has approximately 60K parameters, requires 19M multiply-accumulate operations and improves the performance by approximately 2-4 percentage points compared to the DCASE 2022 Task 1 baseline network.