Optimised soundprint selection for identification in audio streams

An improved iterative method for selection of reference soundprints from a set of sound records to be identified is presented. The soundprints applied for reliable streaming audio identification should have maximum dissimilar ity to each other. As the first trial of the iteration, a set of soundprint s is randomly selected according to a grid placed on the records. To the ac cepted set., a further local refinement process is applied using the quadra tic local approximations of the Euclidean distance between the soundprints. A large number of measurements have shown that the method significantly im proves the reliability of real-time monitoring.