 |
Conference papers
A weakly-supervised discriminative model for audio-to-score alignment
Abstract : In this paper, we consider a new discriminative approach to the problem of audio-to-score alignment. We consider the two distinct informations provided by the music scores: (i) an exact ordered list of musical events and (ii) an approximate prior information about relative duration of events. We extend the basic dynamic time warping algorithm to a convex problem that learns optimal classifiers for all events while jointly aligning files, using this weak supervision only. We show that the relative duration between events can be easily used as a penalization of our cost function and allows us to drastically improve performances of our approach. We demonstrate the validity of our approach on a large and realistic dataset.
Cited literature [26 references]
https://hal.archives-ouvertes.fr/hal-01251018
Contributor : Philippe Cuvillier <>
Submitted on : Tuesday, January 5, 2016 - 3:12:06 PM
Last modification on : Monday, February 10, 2020 - 6:13:49 PM
Document(s) archivé(s) le : Thursday, April 7, 2016 - 3:26:31 PM
Contributor : Philippe Cuvillier <>
Submitted on : Tuesday, January 5, 2016 - 3:12:06 PM
Last modification on : Monday, February 10, 2020 - 6:13:49 PM
Document(s) archivé(s) le : Thursday, April 7, 2016 - 3:26:31 PM
File
icassp2016.pdf
Files produced by the author(s)
