Aiming to an automatic sound recognizer for radio broadcasting events, a methodology of clustering\nthe audio feature space using the discrimination ability of the audio descriptors as a criterion, is\ninvestigated in this work. From a given and close set of audio events, commonly found in broadcast\nnews transmissions, a large set of audio descriptors is extracted and their data-driven ranking of\nrelevance is clustered, providing a more robust feature selection. The clusters of the feature space are\nfeeding machine learning algorithms implemented as classification models during the experimental\nevaluation. This methodology showed that support vector machines provide significantly good\nresults, considering the achieved accuracy due to their ability of coping well in high dimensionality\nexperimental conditions.
Loading....