A simple model for detection of rare sound events

TL;DR

This paper introduces a simple recurrent model that detects rare sound events by combining utterance-level and frame-level classification losses, utilizing attention mechanisms, and demonstrates competitive results on a standard challenge dataset.

Contribution

The paper presents a novel recurrent model that jointly optimizes utterance and frame-level losses with attention, specifically designed for rare sound event detection.

Findings

Achieved competitive performance on DCASE 2017 Task 2

Effectively combines utterance and frame-level classification

Utilizes attention mechanism for improved detection

Abstract

We propose a simple recurrent model for detecting rare sound events, when the time boundaries of events are available for training. Our model optimizes the combination of an utterance-level loss, which classifies whether an event occurs in an utterance, and a frame-level loss, which classifies whether each frame corresponds to the event when it does occur. The two losses make use of a shared vectorial representation the event, and are connected by an attention mechanism. We demonstrate our model on Task 2 of the DCASE 2017 challenge, and achieve competitive performance.