We are attending ICASSP 2019 in Brighten, UK. Minje will chair a session on source separation and speech enhancement; Sunwoo and Sanna will present their papers about bitwise recurrent neural networks and a database of quality Karaoke singing, respectively.