@techreport{Chauhan2013,
    Author = "Chauhan, Sameer and Phadke, Sharang and Sherland, Christian",
    title = "Event Detection and Classification",
    institution = "DCASE2013 Challenge",
    year = "2013",
    month = "June",
    abstract = "The IEEE AASP Challenge addresses the problem of acoustic event detection and classification in an office environment. Our system performs segmentation and event classification on a continuous stream of acoustic activity in an office using basic feature extraction techniques and a single layer frame-by-frame classifier. We achieve high classification accuracy in noiseless environments, but performance severely deteriorates in noisy environments."
}

@techreport{Diment2013,
    Author = "Diment, Aleksandr and Heittola, Toni and Virtanen, Tuomas",
    title = "Sound Event Detection for Office Live and Office Synthetic {AASP} Challenge",
    institution = "DCASE2013 Challenge",
    year = "2013",
    month = "June",
    abstract = "We present a sound event detection system based on hidden Markov models. The system is evaluated with development material provided in the AASP Challenge on Detection and Classification of Acoustic Scenes and Events. Two approaches using the same basic detection scheme are presented. First one, developed for acoustic scenes with non-overlapping sound events is evaluated with Office Live development dataset. Second one, developed for acoustic scenes with some degree of overlapping sound events is evaluated with Office Synthetic development dataset."
}

@techreport{Gemmeke2013,
    Author = "Gemmeke, Jort F and Vuegen, Lode and Vanrumste, Bart and Van hamme, Hugo",
    title = "An Exemplar-Based {NMF} Approach for Audio Event Detection",
    institution = "DCASE2013 Challenge",
    year = "2013",
    month = "June",
    abstract = "We present a novel, exemplar-based method for audio event detection based on non-negative matrix factorisation (NMF). Building on recent work in noise robust automatic speech recognition, we model events as a linear combination of dictionary atoms, and mixtures as a linear combination of overlapping events. The exemplar based dictionary is created by extracting all available training data, artificially augmented by linear time warping at multiple rates. The method is evaluated on the Office Live and Office Synthetic development datasets released by the AASP Challenge on Detection and Classification of Acoustic Scenes and Events."
}

@techreport{Niessen2013,
    Author = "Niessen, Maria E. and Van Kasteren, Tim L. M and Merentitis, Andreas",
    title = "Hierarchical Sound Event Detection",
    institution = "DCASE2013 Challenge",
    year = "2013",
    month = "June",
    abstract = "Environmental sound recognition in real-world conditions is a particularly challenging topic, since it requires significant efforts on both the feature extraction and the classification modeling parts in order to achieve satisfactory results. In the presented work we propose a multi-tier method that employs best of breed techniques at all relevant tasks; initially feature extraction takes place focusing on a broad range of audio features. Following feature extraction a Hierarchical Hidden Markov Model classifier scheme with explicit modeling of the finishing of the state to better detect transitions is developed. Finally, the best result is achieved when ensemble methods are added on top of the previous scheme. Specifically, a variation of Stacking using a Random Forest and the HHMM as level-1 classifiers and a second instance of HHMM as the metaclassifier is selected. Results indicate that this final method is on one hand able to deliver the best overall performance, as well as explore different tradeoffs between classes and metrics (e.g. emphasize on specific metrics or classes that are of higher importance)."
}

@techreport{Nogueira2013,
    Author = "Nogueira, Waldo and Roma, Gerard and Herrera, Perfecto",
    title = "Automatic Event Classification Using Front End Single Channel Noise Reduction, {MFCC} Features and a Support Vector Machine Classifier",
    institution = "DCASE2013 Challenge",
    year = "2013",
    month = "June",
    abstract = "This submission to the sub-task scene event classification Office Live of the IEEE AASP Challenge: Detection and Classification of Acoustic Scenes and Events uses first a single channel noise reduction to clean stationary background noise, next mfccs are extracted and finally a support-vector machine classifier is used to classify the events. In this short paper the usage of the implementation as well as a short description of the system is explained."
}

@techreport{Schroeder2013,
    Author = "Schröder, Jens and Cauchi, Benjamin and Schädler, Marc René and Moritz, Niko and Adiloglu, Kamil and Anemüller, Jörn and Doclo, Simon and Kollmeier, Birger and Goetze, Stefan",
    title = "Acoustic Event Detection Using Signal Enhancement and Spectro-Temporal Feature Extraction",
    institution = "DCASE2013 Challenge",
    year = "2013",
    month = "June",
    abstract = "In this paper, an acoustic event detection system is proposed. It consists of a noise reduction signal enhancement step based on the noise power spectral density estimator proposed in [1] and on the noise suppression by [2], a Gabor filterbank feature extraction stage and a two layer hidden Markov model as back-end classifier. Optimization on the development set yields up to a F-Score of 0.73 on frame based and 0.63 on onset and offset based measure."
}

@techreport{Vuegen2013,
    Author = "Vuegen, Lode and Van Den Broeck, Bert and Karsmakers, Peter and Gemmeke, Jort F and Vanrumste, Bart and Van hamme, Hugo",
    title = "An {MFCC}-{GMM} Approach for Event Detection and Classification",
    institution = "DCASE2013 Challenge",
    year = "2013",
    month = "June",
    abstract = "This abstract explores Gaussian Mixture Models (GMM) estimated from Mel Frequency Cepstral Coefficients (MFCCs) for acoustic event detection and classification. To limit the impact of silence, a shared background model is used. An average Fscore of 48\% for the office life subtask is obtained. However, the analysis reveals that the proposed method has difficulties to cope with the large intra-class variations (e.g. time durations, dynamic range, characteristic sounds) in the provided dataset."
}