Mining Labeled Data from Web-Scale Collections for Vocal Activity Detection in Music

Abstract

This work demonstrates an approach to generating strongly labeled data for vocal activity detection by pairing instrumental versions of songs with their original mixes. Though such pairs are rare, we find ample instances in a massive music collection for training deep convolutional networks at this task, achieving state of the art performance with a fraction of the human effort required previously. Our error analysis reveals two notable insights: imperfect systems may exhibit better temporal precision than human annotators, and should be used to accelerate annotation; and, machine learning from mined data can reveal subtle biases in the data source, leading to a better understanding of the problem itself. We also discuss future directions for the design and evolution of benchmarking datasets to rigorously evaluate AI systems.

Related

July 2020 | IJCAI - International Joint Conference on Artificial Intelligence

Seq-U-Net: A One-Dimensional Causal U-Net for Efficient Sequence Modelling

Daniel Stoller, Mi Tian, Sebastian Ewert, and Simon Dixon

July 2020 | WCCI/IJCNN - IEEE World Congress on Computational Intelligence / International Joint Conference on Neural Networks

Using a Neural Network Codec Approximation Loss to Improve Source Separation Performance in Limited Capacity Networks

Ishwarya Ananthabhotla, Sebastian Ewert, Joseph A. Paradiso