Four papers accepted in ICASSP 2021!

They include our work on Contrastive Learning of Sound Event Representations, as well as other collaborations with other Google and DCASE researchers

FSD50K dataset and paper are out!

Releasing FSD50K dataset, the new open dataset for sound event research with over 100h of audio

Paper on missing labels in AudioSet accepted in IEEE SPL!

Paper is titled "Addressing Missing Labels in Large-scale Sound Event Recognition using a Teacher-student Framework with Loss Masking"

My ICASSP 2020 highlights

My personal highlights from ICASSP 2020