Acoustic event detection based on non-negative matrix factorization with mixtures of local dictionaries and activation aggregation

• Published in: 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) pp. 2259 - 2263
• Main Authors: Komatsu, Tatsuya, Senda, Yuzo, Kondo, Reishi
• Format: Conference Proceeding Journal Article
• Language: English
• Published: IEEE 01.03.2016
• Subjects: Acoustic Event Detection; Acoustics; Activation; activation aggregation; Agglomeration; Conferences; convex cone; Dictionaries; Electronics; Event detection; Factorization; group sparsity; Indexes; NMF; Redundancy; Spectrogram; Speech; Support vector machines
• ISSN: 2379-190X
• DOI: 10.1109/ICASSP.2016.7472079

This paper proposes a new non-negative matrix factorization (NMF) based acoustic event detection (AED) method with mixtures of local dictionaries (MLD) and activation aggregation. One of the key problems of conventional NMF-based methods is instability of activations due to redundancy of a region spanned by the bases of dictionaries. Sounds inside the redundant region are often decomposed into undesired combinations of bases and activations that cause failure of detection. The proposed method employs MLD for allocating sub-groups of basis dictionaries to acoustic elements to minimize redundancy in the region and obtain controlled activations. In order to make activations more stable, the proposed method also introduces activation aggregation which combines basis-wise activations into acoustic-element-wise activations. Much more stable activations by the proposed method lead to significant improvement in F-measure by up to 60% compared to an ordinary convolutive-NMF-based method. The proposed method also outperforms a latest alternative which is not based on NMF.