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

facebooktwittermailshare

MUSIC BOUNDARY DETECTION BASED ON A HYBRID DEEP MODEL OF NOVELTY, HOMOGENEITY, REPETITION AND DURATION

Abstract: 

Current state-of-the-art music boundary detection methods use local features for boundary detection, but such an approach fails to explicitly incorporate the statistical properties of the detected segments. This paper presents a music boundary detection method that simultaneously considers a fitness measure based on the boundary posterior probability, the likelihood of the segmentation duration sequence, and the acoustic consistency within a segment. Evaluation shows that our method improves segmentation F0.58-measure by about 10 points compared to DNN with peak-picking, a popular scheme used in the state-of-the-art music boundary detectors.

up
0 users have voted:

Paper Details

Authors:
Akira Maezawa
Submitted On:
15 May 2019 - 11:50am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Akira Maezawa
Paper Code:
AASP-L7.6
Document Year:
2019
Cite

Document Files

ICASSP2019-maezawa.pdf

(19)

Subscribe

[1] Akira Maezawa, "MUSIC BOUNDARY DETECTION BASED ON A HYBRID DEEP MODEL OF NOVELTY, HOMOGENEITY, REPETITION AND DURATION", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4527. Accessed: Jul. 19, 2019.
@article{4527-19,
url = {http://sigport.org/4527},
author = {Akira Maezawa },
publisher = {IEEE SigPort},
title = {MUSIC BOUNDARY DETECTION BASED ON A HYBRID DEEP MODEL OF NOVELTY, HOMOGENEITY, REPETITION AND DURATION},
year = {2019} }
TY - EJOUR
T1 - MUSIC BOUNDARY DETECTION BASED ON A HYBRID DEEP MODEL OF NOVELTY, HOMOGENEITY, REPETITION AND DURATION
AU - Akira Maezawa
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4527
ER -
Akira Maezawa. (2019). MUSIC BOUNDARY DETECTION BASED ON A HYBRID DEEP MODEL OF NOVELTY, HOMOGENEITY, REPETITION AND DURATION. IEEE SigPort. http://sigport.org/4527
Akira Maezawa, 2019. MUSIC BOUNDARY DETECTION BASED ON A HYBRID DEEP MODEL OF NOVELTY, HOMOGENEITY, REPETITION AND DURATION. Available at: http://sigport.org/4527.
Akira Maezawa. (2019). "MUSIC BOUNDARY DETECTION BASED ON A HYBRID DEEP MODEL OF NOVELTY, HOMOGENEITY, REPETITION AND DURATION." Web.
1. Akira Maezawa. MUSIC BOUNDARY DETECTION BASED ON A HYBRID DEEP MODEL OF NOVELTY, HOMOGENEITY, REPETITION AND DURATION [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4527