Environment Sound Classification using Multiple Feature Channels and Attention based Deep Convolutional Neural Network

TL;DR

This paper introduces a novel deep learning model for environment sound classification that uses multiple feature channels and attention mechanisms, achieving state-of-the-art results across multiple benchmark datasets.

Contribution

The paper presents the first use of multiple feature channels including MFCC, GFCC, CQT, and Chromagram in a deep CNN with attention for sound classification, surpassing previous models.

Findings

Achieved state-of-the-art accuracy on UrbanSound8K, ESC-10, and ESC-50 datasets.

Model accuracy exceeds human performance on ESC-10 and ESC-50.

Introduced a deeper CNN with spatially separable convolutions and combined channel and spatial attention.

Abstract

In this paper, we propose a model for the Environment Sound Classification Task (ESC) that consists of multiple feature channels given as input to a Deep Convolutional Neural Network (CNN) with Attention mechanism. The novelty of the paper lies in using multiple feature channels consisting of Mel-Frequency Cepstral Coefficients (MFCC), Gammatone Frequency Cepstral Coefficients (GFCC), the Constant Q-transform (CQT) and Chromagram. Such multiple features have never been used before for signal or audio processing. And, we employ a deeper CNN (DCNN) compared to previous models, consisting of spatially separable convolutions working on time and feature domain separately. Alongside, we use attention modules that perform channel and spatial attention together. We use some data augmentation techniques to further boost performance. Our model is able to achieve state-of-the-art performance on all three benchmark environment sound classification datasets, i.e. the UrbanSound8K (97.52%), ESC-10 (95.75%) and ESC-50 (88.50%). To the best of our knowledge, this is the first time that a single environment sound classification model is able to achieve state-of-the-art results on all three datasets. For ESC-10 and ESC-50 datasets, the accuracy achieved by the proposed model is beyond human accuracy of 95.7% and 81.3% respectively.