MuSE::Spatial structure in audio representations

​Research talk by Chuyang Chen, PhD student, NYU MARL, on "Probing Spatial Structure in Pretrained Audio Representations" (Chen, Ding, Roman, Bello). Pretrained audio models know what a sound is. Do they know where it came from, or what room it was in? Chuyang presents SARL, a benchmark that puts that question to thirteen widely used models. No background needed. Talk and Q&A, open to all NYU students.

​For those who aren’t NYU-affiliates or won’t be able to make it in person, please join us online:

​https://nyu.zoom.us/j/96543996147