Data-driven audio recognition: a supervised dictionary approach

TL;DR

This paper introduces a supervised dictionary learning method for machine hearing, demonstrating its effectiveness across auditory scene analysis and music chord recognition by achieving state-of-the-art results.

Contribution

The paper presents a novel, efficient supervised dictionary learning approach that advances data-driven audio recognition methods.

Findings

Achieves state-of-the-art performance on auditory scene datasets.

Effective in synthetic music chord recognition.

Demonstrates versatility across different audio recognition tasks.

Abstract

Machine hearing is an emerging area. Motivated by the need of a principled framework across domain applications for machine listening, we propose a generic and data-driven representation learning approach. For this sake, a novel and efficient supervised dictionary learning method is presented. Experiments are performed on both computational auditory scene (East Anglia and Rouen) and synthetic music chord recognition datasets. Obtained results show that our method is capable to reach state-of-the-art hand-crafted features for both applications