English

A Music Classification Model based on Metric Learning and Feature Extraction from MP3 Audio Files

Sound 2019-09-19 v2 Information Retrieval Machine Learning Audio and Speech Processing Machine Learning

Abstract

The development of models for learning music similarity and feature extraction from audio media files is an increasingly important task for the entertainment industry. This work proposes a novel music classification model based on metric learning and feature extraction from MP3 audio files. The metric learning process considers the learning of a set of parameterized distances employing a structured prediction approach from a set of MP3 audio files containing several music genres. The main objective of this work is to make possible learning a personalized metric for each customer. To extract the acoustic information we use the Mel-Frequency Cepstral Coefficient (MFCC) and make a dimensionality reduction with the use of Principal Components Analysis. We attest the model validity performing a set of experiments and comparing the training and testing results with baseline algorithms, such as K-means and Soft Margin Linear Support Vector Machine (SVM). Experiments show promising results and encourage the future development of an online version of the learning model.

Keywords

Cite

@article{arxiv.1905.12804,
  title  = {A Music Classification Model based on Metric Learning and Feature Extraction from MP3 Audio Files},
  author = {Angelo C. Mendes da Silva and Mauricio A. Nunes and Raul Fonseca Neto},
  journal= {arXiv preprint arXiv:1905.12804},
  year   = {2019}
}

Comments

In a review process, I found some errors and made some changes in methodology that improved my results. Once I finish the experiments, I will upload the new version

R2 v1 2026-06-23T09:32:32.094Z