General-purpose Tagging of Freesound Audio with AudioSet Labels: Task Description, Dataset, and Baseline
This is an incremental contribution, providing a standardized benchmark for audio tagging research using Freesound and AudioSet labels.
The paper describes a 2018 challenge task for building an audio tagging system to recognize 41 categories from audio clips, and presents the task, dataset, and a baseline system.
This paper describes Task 2 of the DCASE 2018 Challenge, titled "General-purpose audio tagging of Freesound content with AudioSet labels". This task was hosted on the Kaggle platform as "Freesound General-Purpose Audio Tagging Challenge". The goal of the task is to build an audio tagging system that can recognize the category of an audio clip from a subset of 41 diverse categories drawn from the AudioSet Ontology. We present the task, the dataset prepared for the competition, and a baseline system.