Coupled Sparse NMF vs. Random Forest Classification for Real Life Acoustic Event Detection

Sobieraj, Iwona; Plumbley, Mark D.

by Iwona Sobieraj, Mark D. Plumbley

Abstract:

Coupled non-negative matrix factorization (NMF) of spectral representations and class activity annotations has shown promising results for acoustic event detection (AED) in real life environments. Recently, a new dataset has been proposed for development of algorithms for real life AED. In this paper we propose two methods for real life polyphonic AED: Coupled Sparse Non-negative Matrix Factorization (CSNMF) of time-frequency patches with class activity annotations and Multi-class Random Forest classification (MRF) of time-frequency patches, and compare their performance on this new dataset. Both our methods outperform the DCASE2016 baseline in terms of F-score. Moreover, we show that as the dataset is unbalanced, a classifier that recognizes a few most frequent classes may outperform the sparse NMF-approach and a baseline based on Gaussian Mixture Models.

Reference:

Iwona Sobieraj, Mark D. Plumbley, "Coupled Sparse NMF vs. Random Forest Classification for Real Life Acoustic Event Detection", In Proceedings of the Detection and Classification of Acoustic Scenes and Events 2016 Workshop (DCASE2016), pp. 90-94, 2016.

Bibtex Entry:

@inproceedings{2016_Sobieraj_Coupled,
    Author = "Sobieraj, Iwona and Plumbley, Mark D.",
    title = "Coupled Sparse {NMF} vs. Random Forest Classification for Real Life Acoustic Event Detection",
    booktitle = "Proceedings of the Detection and Classification of Acoustic Scenes and Events 2016 Workshop (DCASE2016)",
    month = "September",
    pages = "90--94",
    year = "2016",
    keywords = "Acoustic event detection, random forest classifier, non-negative matrix factorization, sparse representation",
    abstract = "Coupled non-negative matrix factorization (NMF) of spectral representations and class activity annotations has shown promising results for acoustic event detection (AED) in real life environments. Recently, a new dataset has been proposed for development of algorithms for real life AED. In this paper we propose two methods for real life polyphonic AED: Coupled Sparse Non-negative Matrix Factorization (CSNMF) of time-frequency patches with class activity annotations and Multi-class Random Forest classification (MRF) of time-frequency patches, and compare their performance on this new dataset. Both our methods outperform the DCASE2016 baseline in terms of F-score. Moreover, we show that as the dataset is unbalanced, a classifier that recognizes a few most frequent classes may outperform the sparse NMF-approach and a baseline based on Gaussian Mixture Models."
}