Learning Contextualized Music Semantics from Tags via a Siamese Network

TL;DR

This paper presents a Siamese neural network approach that models contextualized music semantics from tags, improving music tag understanding and addressing out-of-vocabulary issues through unsupervised learning.

Contribution

It introduces a novel Siamese network-based method that captures contextualized music semantics using tag features and probabilistic topic models, outperforming existing methods.

Findings

Outperforms state-of-the-art in semantic priming

Enhances music tag completion accuracy

Addresses out-of-vocabulary problems effectively

Abstract

Music information retrieval faces a challenge in modeling contextualized musical concepts formulated by a set of co-occurring tags. In this paper, we investigate the suitability of our recently proposed approach based on a Siamese neural network in fighting off this challenge. By means of tag features and probabilistic topic models, the network captures contextualized semantics from tags via unsupervised learning. This leads to a distributed semantics space and a potential solution to the out of vocabulary problem which has yet to be sufficiently addressed. We explore the nature of the resultant music-based semantics and address computational needs. We conduct experiments on three public music tag collections -namely, CAL500, MagTag5K and Million Song Dataset- and compare our approach to a number of state-of-the-art semantics learning approaches. Comparative results suggest that this approach outperforms previous approaches in terms of semantic priming and music tag completion.