Organizer of Task 4 “Large-scale weakly labeled semi-supervised sound event detection in domestic environments”. Accountable for code development, dataset development, paper reviews and system submissions as well as providing technical support to participants via email and DCASE forum.
The task evaluates systems for the large-scale detection of sound events using weakly labeled data. The challenge is to explore the possibility to exploit a large amount of unbalanced and unlabelled training data together with a small weakly annotated training set to improve system performance. The data are YouTube video excerpts focusing on domestic context which could be used for example in ambient assisted living applications. The domain was chosen due to the scientific challenges (wide variety of sounds, time-localized events…) and potential industrial applications.