A Low-Compexity Deep Learning Framework For Acoustic Scene Classification

TL;DR

This paper introduces a low-complexity deep learning framework for acoustic scene classification that combines multiple spectrogram types, CNNs, and model compression techniques to achieve high accuracy with minimal parameters.

Contribution

The paper presents a novel low-complexity CNN framework using multiple spectrograms and model compression for efficient acoustic scene classification.

Findings

Achieved 66.7% classification accuracy on DCASE 2021 dataset.

Reduced model size to 128 KB with model restriction and decomposed convolution.

Improved baseline accuracy by 19%.

Abstract

In this paper, we presents a low-complexity deep learning frameworks for acoustic scene classification (ASC). The proposed framework can be separated into three main steps: Front-end spectrogram extraction, back-end classification, and late fusion of predicted probabilities. First, we use Mel filter, Gammatone filter and Constant Q Transfrom (CQT) to transform raw audio signal into spectrograms, where both frequency and temporal features are presented. Three spectrograms are then fed into three individual back-end convolutional neural networks (CNNs), classifying into ten urban scenes. Finally, a late fusion of three predicted probabilities obtained from three CNNs is conducted to achieve the final classification result. To reduce the complexity of our proposed CNN network, we apply two model compression techniques: model restriction and decomposed convolution. Our extensive experiments, which are conducted on DCASE 2021 (IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events) Task 1A development dataset, achieve a low-complexity CNN based framework with 128 KB trainable parameters and the best classification accuracy of 66.7%, improving DCASE baseline by 19.0%