Cascade classifiers trained on gammatonegrams for reliably detecting audio events

Reliable detection of audio events in highly noisy environments

“…The temporal sequence of symbols that represent spectral shapes has been taken into account by Chin and Burred [7], that classify the audio events by matching sub-sequences of the reference events using the Genetic Motif Discovery technique. The event detection task is formulated by Foggia et al [16] as an object detection problem in the Gammatone image of the sound.…”

Trainable COPE Features for Sound Event Detection

“…energy, pitch, bandwidth, etc. Feature vectors were then used to train a classifier model, subsequently used in the operating phase for the detection and classification of events of interest in unknown audio streams [6,11,14,23,30]. Subsequently, more complex architectures or data representations were proposed to strengthen the robustness of the system to high variability of the background noise: multi-stage classifiers [19], a classification rejection module [7], representations of the audio signal based on the bag of features approach [12,20].…”

“…The temporal arrangement of instantaneous features was taken into account in [15], and in [22] with a pyramidal approach. In [11], instead the audio stream was represented as a time-frequency distribution of its energy and object-detection techniques were employed to detect the events of interest. All these methods involve a feature engineering step, in which a proper set of features for the problem at hand is defined.…”

Automatic Detection and Classification of Audio Events for Road Surveillance Applications

“…The TS event is considered as an abnormal event because a frequent occurrence of this event is a sign of a dangerous and busy road state, which may require attention from traffic personnel to ensure safety. The problem of audio events detection and classification such as gunshots, explosions, and screams has been addressed in several previous studies [ 15 , 16 , 22 , 23 ]. Although various audio surveillance system setups and additional audio features have been explored, most of the approaches proposed in the previous studies are based on the conventional methods of modeling short-term power spectrum features such as the Mel-frequency cepstral coefficient (MFCC) [ 24 ], using either Gaussian mixture models (GMMs) or Hidden Markov models (HMMs) [ 20 ].…”

“…The proposed technique employs a short-time analysis based on features observed in the temporal, spectral and the joint time–frequency ( t , f ) domain, extracted from quadratic time–frequency distributions (QTFDs), for sound detection and classification. Whereas previous studies have either used a combination of temporal and spectral features [ 13 , 17 ] or ( t , f )-based techniques [ 22 , 23 ] alone for non-stationary signal classification such as audio data, this work is novel in the sense that we have used a combination of t -, f - and ( t , f )-domain features. In addition, we have extracted ( t , f )-features from high resolution and robust QTFDs, whereas in previous studies time-frequency features are extracted from conventional spectrograms.…”