FSDKaggle2018 Audio Classification Dataset

FSDKaggle2018 is an audio dataset that can be used for tasks such as audio annotation and sound source separation. The dataset contains 11,073 audio files and is annotated with 41 categories of AudioSet Ontology.

This dataset has been used for DCASE Challenge 2018 Task 2. All audio samples are from the Freesound dataset and are provided as uncompressed PCM 16 bit, 44.1 kHz mono audio files.