Learning Sparse Adversarial Dictionaries For Multi-Class Audio Classification
This work addresses audio classification challenges for applications like denoising and speech enhancement, but it is incremental as it builds on existing sparse dictionary methods.
The paper tackled the problem of overlapping and noisy audio events by modifying traditional dictionary learning to incorporate a discriminative term, learning class-specific adversarial dictionaries that improve representation for their own class while hindering others, achieving effective performance as a stand-alone solution for binary and multi-class audio classification.
Audio events are quite often overlapping in nature, and more prone to noise than visual signals. There has been increasing evidence for the superior performance of representations learned using sparse dictionaries for applications like audio denoising and speech enhancement. This paper concentrates on modifying the traditional reconstructive dictionary learning algorithms, by incorporating a discriminative term into the objective function in order to learn class-specific adversarial dictionaries that are good at representing samples of their own class at the same time poor at representing samples belonging to any other class. We quantitatively demonstrate the effectiveness of our learned dictionaries as a stand-alone solution for both binary as well as multi-class audio classification problems.