Acoustic Scene Classification: A Competition Review

TL;DR

The methods and results discovered during a competition organized in the context of a graduate machine learning course are described and its importance in the curriculum based on student feedback is justified.

Abstract

In this paper we study the problem of acoustic scene classification, i.e., categorization of audio sequences into mutually exclusive classes based on their spectral content. We describe the methods and results discovered during a competition organized in the context of a graduate machine learning course; both by the students and external participants. We identify the most suitable methods and study the impact of each by performing an ablation study of the mixture of approaches. We also compare the results with a neural network baseline, and show the improvement over that. Finally, we discuss the impact of using a competition as a part of a university course, and justify its importance in the curriculum based on student feedback.

Figures (2)

• Figure 1: Illustration of the effects of Random Erasing.
• Figure 2: Illustration of mixup augmentation. Left: Two original mel-spectrogram samples from the training set. Right: two mixtures of the original samples.