Large-scale audio event discovery in one million YouTube videos

37 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

FSD50K: An Open Dataset of Human-Labeled Sound Events (2020, background): “1 SER has gained increasing attention in the past few years, becoming a key component in applications related to healthcare [1]–[3], urban sound planning [4], bioacoustics monitoring [5]–[7], multimedia event detection [8] and large-scale event discovery [9], surveillance [10, 11], or noise monitori”
Coincidence, Categorization, and Consolidation: Learning to Recognize Sounds with Minimal Supervision (2019, background): “Cluster-based active learning in ﬁxed representation spaces has been proposed in past research [14, 15].”
Search Result Clustering in Collaborative Sound Collections (2020, background): “neural networks often learn similar features which can be applicable for many datasets and tasks [45]. Intermediate layers can serve as a higher-level representations which can be used for clustering [22]. Clustering is a type of unsupervised classification which consists in organising similar objec”
Look, Listen, and Learn More: Design Choices for Deep Audio Embeddings (2019, methodology): “Machine listening is an active area of research concerned with the development of computational methods to derive meaning from sound, in which the use of deep learning has seen growing popularity and success [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13].”