A regressive boosting approach to automatic audio tagging based on soft annotator fusion

Abstract : Automatic tagging of music has mostly been treated as a classification problem. In this framework, the association of a tag to a song is characterized in a " hard " fashion: the tag is either relevant or not. Yet, the relevance of a tag to a song is not always evident. Indeed, during the ground-truth annotation process, several annotators may express doubts, or disagree with each other. In this paper, we propose to fuse annota-tors' decisions in a way to keep information about this uncertainty. This fusion provides us continuous scores, that are used for training a regressive boosting algorithm. Our experiments show that regression with this soft ground truth leads to a more accurate learning, and better predictions, compared to traditionally used binary classification.
Document type :
Conference papers
Complete list of metadatas

Cited literature [13 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01132529
Contributor : Mathieu Lagrange <>
Submitted on : Tuesday, March 17, 2015 - 2:43:00 PM
Last modification on : Thursday, October 17, 2019 - 12:36:09 PM
Long-term archiving on : Monday, April 17, 2017 - 4:40:15 PM

File

RF_ICASSP-12.pdf
Files produced by the author(s)

Identifiers

Citation

Rémi Foucard, Slim Essid, Mathieu Lagrange, Gaël Richard. A regressive boosting approach to automatic audio tagging based on soft annotator fusion. IEEE ICASSP, Mar 2012, Kyoto, Japan. ⟨10.1109/ICASSP.2012.6287820⟩. ⟨hal-01132529⟩

Share

Metrics

Record views

634

Files downloads

220