VOICe: A novel dataset for the development and evaluation of generalizable sound event detection domain adaptation methods! VOICe consists of 1449 different mixtures of three different sound events ("baby crying", "glass breaking", and "gunshot"): 1242 mixtures with background noise of three different categories of acoustic scenes ("vehicle"," outdoors", and "indoors"), mixed under 2 SNR values (-3, -9 dB), that is 207 mixtures x 3 acoustic scenes x 2 SNRs = 1242 207 mixtures without any background noise. VOICe is offered for sound event detection domain adaptation from one acoustic scene to another, or between sound events with background noise and without background noise. You can also find more information about the dataset in our paper: https://arxiv.org/pdf/1911.07098.pdf