Label Tree Embeddings for Acoustic Scene Classification
Multimedia
2016-07-27 v2 Artificial Intelligence
Sound
Abstract
We present in this paper an efficient approach for acoustic scene classification by exploring the structure of class labels. Given a set of class labels, a category taxonomy is automatically learned by collectively optimizing a clustering of the labels into multiple meta-classes in a tree structure. An acoustic scene instance is then embedded into a low-dimensional feature representation which consists of the likelihoods that it belongs to the meta-classes. We demonstrate state-of-the-art results on two different datasets for the acoustic scene classification task, including the DCASE 2013 and LITIS Rouen datasets.
Cite
@article{arxiv.1606.07908,
title = {Label Tree Embeddings for Acoustic Scene Classification},
author = {Huy Phan and Lars Hertel and Marco Maass and Philipp Koch and Alfred Mertins},
journal= {arXiv preprint arXiv:1606.07908},
year = {2016}
}
Comments
to appear in the Proceedings of ACM Multimedia 2016 (ACMMM 2016)