Real-world acoustic event detection

Feature extraction based on the high-pass filtering of audio signals for Acoustic Event Classification

“…The composition of the whole database was intended to be similar to the one used in (Zhuang et al, 2010) and it is shown in Table 1. Audio files were obtained from different sources: websites 2 , the FBK-Irst database 3 (FBK-Irst, 2009) and the UPC-TALP database 4 (UPC-TALP, 2008) and were converted to the same format and sampling frequency (8 kHz).…”

“…Several front-ends have been proposed in the literature, some of them based on short-term features, such as Mel-Frequency Cepstral Coefficients (MFCC) (Temko and Nadeu, 2006;Zieger, 2008;Zhuang et al, 2010;Kwangyoun and Hanseok, 2011), log filterbank energies (Zhuang et al, 2010), Perceptual Linear Prediction (PLP) (Portelo et al, 2009), log-energy, spectral flux, fundamental entropy and zero-crossing rate (Temko and Nadeu, 2006). Other approaches are based on the application of different temporal integration techniques over these short-term features (Meng et al, 2007;Mejía-Navarrete et al, 2011;Zhang and Schuller, 2012).…”

“…In fact, as pointed in (Zhuang et al, 2010), conventional acoustic features are not necessarily the more appropriate for AEC tasks because they have been design according to the spectral characteristics of speech which are quite different from the spectral structure of acoustic events. To deal with this issue in (Zhuang et al, 2010) it is proposed a boosted feature selection method to construct a more suitable parameterization for AEC.…”

“…To deal with this issue in (Zhuang et al, 2010) it is proposed a boosted feature selection method to construct a more suitable parameterization for AEC.…”

Detection and localization of selected acoustic events in acoustic field for smart surveillance applications

“…The known solutions for localization of acoustic events in most cases use a microphone (pressure sensor) array and are limited to the calculation of acoustic wave direction of arrival (DoA) [14,17]. In this work a novel approach is introduced employing the acoustic vector sensor (AVS) which makes possible to calculate not only the direction of arrival, but also the exact position of the sound source in 3D space [4].…”