The NIGENS General Sound Events Database

TL;DR

The paper introduces NIGENS, a comprehensive and well-annotated sound event database with 714 isolated sound clips of 14 types and additional general sounds, supporting research in auditory scene analysis.

Contribution

It provides a new, high-quality, and extensively labeled sound database specifically designed for training and testing sound event detection models.

Findings

Contains 714 high-quality sound event clips of 14 types

Includes 303 general sound files with diverse audio content

Features precise annotations with perceptual on- and offset times

Abstract

Computational auditory scene analysis is gaining interest in the last years. Trailing behind the more mature field of speech recognition, it is particularly general sound event detection that is attracting increasing attention. Crucial for training and testing reasonable models is having available enough suitable data -- until recently, general sound event databases were hardly found. We release and present a database with 714 wav files containing isolated high quality sound events of 14 different types, plus 303 `general' wav files of anything else but these 14 types. All sound events are strongly labeled with perceptual on- and offset times, paying attention to omitting in-between silences. The amount of isolated sound events, the quality of annotations, and the particular general sound class distinguish NIGENS from other databases.