caa a22 4500 605447306 CHVBK 20210128100131.0 cr unu---uuuuu 210128e20150701xx s 000 0 eng 10.1007/s11042-013-1842-9 doi (NATIONALLICENCE)springer-10.1007/s11042-013-1842-9 Multi-label audio concept detection using correlated-aspect Gaussian Mixture Model [Elektronische Daten] [Cencen Zhong, Zhenjiang Miao] As an essentially multi-label classification problem, audio concept detection is normally solved by treating concepts independently. Since in this process the original useful concept correlation information is missing, this paper proposes a new model named Correlated-Aspect Gaussian Mixture Model (C-AGMM) to take advantage of such a clue for enhancing multi-label audio concept detection. Originating from Aspect Gaussian Mixture Model (AGMM) which improves GMM by incorporating it into probabilistic Latent Semantic Analysis (pLSA), C-AGMM still learns a probabilistic model of the whole audio clip by regarding concepts as its component elements. However, different from AGMM that assumes concepts independent with each other, C-AGMM considers their distribution on a sub-manifold embedded in the ambient space. With an assumption that if two concepts are close in the intrinsic geometry of this distribution then their conditional probability distributions are likely to show similarity, a graph regularizer is exploited to model the correlation between these concepts. Following the Maximum Likelihood Estimate principle, model parameters of C-AGMM encoding the concept correlation clue are derived and used directly as the detection criterion. Experiments on two datasets show the effectiveness of our proposed model. Springer Science+Business Media New York, 2014 Aspect Gaussian Mixture Model nationallicence Probabilistic latent semantic analysis nationallicence Audio concept detection nationallicence Concept correlation nationallicence Multi-label classification nationallicence Zhong Cencen Institute of Information Science, Beijing Jiaotong University, Beijing, China aut Miao Zhenjiang Institute of Information Science, Beijing Jiaotong University, Beijing, China aut Multimedia Tools and Applications Springer US; http://www.springer-ny.com 74/13(2015-07-01), 4817-4832 1380-7501 74:13<4817 2015 74 11042 https://doi.org/10.1007/s11042-013-1842-9 text/html Onlinezugriff via DOI BK010053 XK010053 XK010000 Metadata rights reserved Springer special CC-BY-NC licence nationallicence 1 research-article jats NATIONALLICENCE NATIONALLICENCE NL-springer NATIONALLICENCE 856 40 https://doi.org/10.1007/s11042-013-1842-9 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Zhong Cencen Institute of Information Science, Beijing Jiaotong University, Beijing, China aut NATIONALLICENCE 700 1- Miao Zhenjiang Institute of Information Science, Beijing Jiaotong University, Beijing, China aut NATIONALLICENCE 773 0- Multimedia Tools and Applications Springer US; http://www.springer-ny.com 74/13(2015-07-01), 4817-4832 1380-7501 74:13<4817 2015 74 11042