Hi all, the slides and notes for Tasha's talk yesterday are now up on the
group page. Our next talk will be on October 21st, at 4pm in CEPSR 620. We're hosting Brian McFee, previously a postdoc at Columbia and now at NYU. The talk abstract follows. Please distribute to anyone you think would be interested! See you next Wednesday.
Musical Data Augmentation
Brian McFee, NYU
October 21st, 4pm, CEPSR 620
Predictive models for music annotation tasks are practically limited by a paucity of well-annotated training data. In the broader context of large-scale machine learning, the concept of "data augmentation" - supplementing a training set with carefully perturbed samples - has emerged as an important component of robust systems. In this work, we develop a general software framework for augmenting annotated musical data sets, which will allow practitioners to easily expand training sets with musically motivated perturbations of both audio and annotations. As a proof of concept, we investigate the effects of data augmentation on the task of recognizing instruments in mixed signals.
∿