A sound source classification system based on subband processing

A classification system that aims to recognize the presence of sounds from different sources is described. The type of audio signals considered are speech, music, noise and silence. Appropriate subband processing is applied for the characterization of each sound source. The algorithm operates in four steps to classify the contents of a given audio signal. The acoustical parameters and statistical measures to be used in the classification process are obtained via an off-line training procedure. In the silence and onset detection stages, we aim to label the starting and finishing instants of the acoustical events present in the audio signal. Acoustical parameters of the given signal are extracted and classification is carried out using linear discrimination with common covariance matrix. Experimental work is carried out on a database that contains mixtures of human speech, musical instruments, background noise of different types and silence. Experimental results demonstrate that the system yields %XX.X classification success for speech/music mixtures, %XX.X for speech/noise mixtures, %XX.X for different musical instruments, %XX.X for mixtures containing speech, music and noise.