Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Deep ranking: triplet matchnet for music metric learning

Abstract: 

Metric learning for music is an important problem for many music information retrieval (MIR) applications such as music generation, analysis, retrieval, classification and recommendation. Traditional music metrics are mostly defined on linear transformations of handcrafted audio features, and may be improper in many situations given the large variety of mu- sic styles and instrumentations. In this paper, we propose a deep neural network named Triplet MatchNet to learn metrics directly from raw audio signals of triplets of music excerpts with human-annotated relative similarity in a supervised fashion. It has the advantage of learning highly nonlinear feature representations and metrics in this end-to-end architecture. Experiments on a widely used music similarity measure dataset show that our method significantly outperforms three state-of-the-art music metric learning methods. Experiments also show that the learned features better preserve the partial orders of the relative similarity than handcrafted features.

up
1 user has voted: Yougen Yuan

Paper Details

Authors:
Rui Lu, Kailun Wu, Zhiyao Duan, Changshui Zhang
Submitted On:
2 March 2017 - 2:56am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Rui Lu
Paper Code:
1205
Document Year:
2017
Cite

Document Files

presentation.pdf

(1288)

Subscribe

[1] Rui Lu, Kailun Wu, Zhiyao Duan, Changshui Zhang, "Deep ranking: triplet matchnet for music metric learning ", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1574. Accessed: Sep. 16, 2019.
@article{1574-17,
url = {http://sigport.org/1574},
author = {Rui Lu; Kailun Wu; Zhiyao Duan; Changshui Zhang },
publisher = {IEEE SigPort},
title = {Deep ranking: triplet matchnet for music metric learning },
year = {2017} }
TY - EJOUR
T1 - Deep ranking: triplet matchnet for music metric learning
AU - Rui Lu; Kailun Wu; Zhiyao Duan; Changshui Zhang
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1574
ER -
Rui Lu, Kailun Wu, Zhiyao Duan, Changshui Zhang. (2017). Deep ranking: triplet matchnet for music metric learning . IEEE SigPort. http://sigport.org/1574
Rui Lu, Kailun Wu, Zhiyao Duan, Changshui Zhang, 2017. Deep ranking: triplet matchnet for music metric learning . Available at: http://sigport.org/1574.
Rui Lu, Kailun Wu, Zhiyao Duan, Changshui Zhang. (2017). "Deep ranking: triplet matchnet for music metric learning ." Web.
1. Rui Lu, Kailun Wu, Zhiyao Duan, Changshui Zhang. Deep ranking: triplet matchnet for music metric learning [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1574