An Open-set Recognition and Few-Shot Learning Dataset for Audio Event Classification in Domestic Environments

TL;DR

This paper introduces a new annotated dataset for open-set recognition and few-shot learning in domestic audio event classification, addressing the lack of dedicated resources and demonstrating baseline results with transfer learning.

Contribution

It provides a carefully annotated dataset for audio FSL in OSR scenarios and benchmarks baseline performance using transfer learning methods.

Findings

The dataset contains 1360 clips from 34 classes, including pattern and unwanted sounds.

Baseline transfer learning models achieve promising results on the dataset.

The dataset facilitates research in audio FSL and OSR in domestic environments.

Abstract

The problem of training with a small set of positive samples is known as few-shot learning (FSL). It is widely known that traditional deep learning (DL) algorithms usually show very good performance when trained with large datasets. However, in many applications, it is not possible to obtain such a high number of samples. In the image domain, typical FSL applications include those related to face recognition. In the audio domain, music fraud or speaker recognition can be clearly benefited from FSL methods. This paper deals with the application of FSL to the detection of specific and intentional acoustic events given by different types of sound alarms, such as door bells or fire alarms, using a limited number of samples. These sounds typically occur in domestic environments where many events corresponding to a wide variety of sound classes take place. Therefore, the detection of such alarms in a practical scenario can be considered an open-set recognition (OSR) problem. To address the lack of a dedicated public dataset for audio FSL, researchers usually make modifications on other available datasets. This paper is aimed at poviding the audio recognition community with a carefully annotated dataset (https://zenodo.org/record/3689288) for FSL in an OSR context comprised of 1360 clips from 34 classes divided into pattern sounds} and unwanted sounds. To facilitate and promote research on this area, results with state-of-the-art baseline systems based on transfer learning are also presented.