This paper proposes innovative de-noise filters in a framework, whose aim is the localization of an acoustic source in a noisy environment. The main focuses are the automatic detection of transient sound events and the separation of the events of interest from the noise. A microphone array is used to capture time-spatial information and an adaptive filter can be initialized to learn the ambient noise spectrum when signals of interest are absent. We propose an algorithm based on the Short Time Spectral Attenuation method to remove the noise from each sensor of the array, before the source localization task is performed. The Time Difference Of Arrival (TDOA) methods are used for multiple sources localization. The experimental results show the efficiency of our framework in stationary noisy environments.
Improvement of acoustic localization using a short Time Spectral Attenuation with a novel suppression rule
Salvati D.;
2009-01-01
Abstract
This paper proposes innovative de-noise filters in a framework, whose aim is the localization of an acoustic source in a noisy environment. The main focuses are the automatic detection of transient sound events and the separation of the events of interest from the noise. A microphone array is used to capture time-spatial information and an adaptive filter can be initialized to learn the ambient noise spectrum when signals of interest are absent. We propose an algorithm based on the Short Time Spectral Attenuation method to remove the noise from each sensor of the array, before the source localization task is performed. The Time Difference Of Arrival (TDOA) methods are used for multiple sources localization. The experimental results show the efficiency of our framework in stationary noisy environments.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
