Deep Convolutional Neural Networks and Data Augmentation for Environmental Sound Classification

TLDR

Deep CNNs can learn discriminative spectro‑temporal patterns, but limited labeled data hinders their use for environmental sound classification. The study aims to develop a deep CNN architecture and evaluate audio data augmentation to address data scarcity in environmental sound classification. The authors design a deep CNN and apply various audio augmentations, training the model on augmented datasets to assess their impact. With augmentation, the proposed deep CNN achieves state‑of‑the‑art performance, outperforming both the unaugmented CNN and a shallow dictionary‑learning model, and revealing class‑specific augmentation effects that suggest further gains with class‑conditional augmentation.

Abstract

The ability of deep convolutional neural networks (CNNs) to learn discriminative spectro-temporal patterns makes them well suited to environmental sound classification. However, the relative scarcity of labeled data has impeded the exploitation of this family of high-capacity models. This study has two primary contributions: first, we propose a deep CNN architecture for environmental sound classification. Second, we propose the use of audio data augmentation for overcoming the problem of data scarcity and explore the influence of different augmentations on the performance of the proposed CNN architecture. Combined with data augmentation, the proposed model produces state-of-the-art results for environmental sound classification. We show that the improved performance stems from the combination of a deep, high-capacity model and an augmented training set: this combination outperforms both the proposed CNN without augmentation and a "shallow" dictionary learning model with augmentation. Finally, we examine the influence of each augmentation on the model's classification accuracy for each class, and observe that the accuracy for each class is influenced differently by each augmentation, suggesting that the performance of the model could be improved further by applying class-conditional data augmentation.