Metric Learning with Background Noise Class for Few-shot Detection of Rare Sound Events

TL;DR

This paper introduces a metric learning approach that explicitly models background noise as a separate class to improve few-shot detection of rare sound events in complex audio sequences, outperforming previous methods.

Contribution

It presents a novel metric learning framework that incorporates background noise as an independent class with a specialized loss and sampling strategy for better separation.

Findings

Outperforms metric learning without background noise class

Achieves detection performance comparable to large-data baseline

Effective in complex audio environments with background noise

Abstract

Few-shot learning systems for sound event recognition have gained interests since they require only a few examples to adapt to new target classes without fine-tuning. However, such systems have only been applied to chunks of sounds for classification or verification. In this paper, we aim to achieve few-shot detection of rare sound events, from query sequence that contain not only the target events but also the other events and background noise. Therefore, it is required to prevent false positive reactions to both the other events and background noise. We propose metric learning with background noise class for the few-shot detection. The contribution is to present the explicit inclusion of background noise as an independent class, a suitable loss function that emphasizes this additional class, and a corresponding sampling strategy that assists training. It provides a feature space where the event classes and the background noise class are sufficiently separated. Evaluations on few-shot detection tasks, using DCASE 2017 task2 and ESC-50, show that our proposed method outperforms metric learning without considering the background noise class. The few-shot detection performance is also comparable to that of the DCASE 2017 task2 baseline system, which requires huge amount of annotated audio data.