Query by example of multimedia signals aims at automatic retrieval of media samples from a database, which are similar to a userprovided example. This paper proposes a method for query by example of audio signals. The method calculates a set of acoustic features from the signals and models their probability density functions (pdfs) using Gaussian mixture models. The method measures the similarity between two samples using the Euclidian distance between their pdfs. A novel method for calculating the closed form solution of the distance is proposed. Simulation experiments show that proposed method enables higher retrieval accuracy than the reference methods.

This paper proposes measures for estimating the similarity of two audio signals, the objective being in query-by-example. Both signals are first represented using a set of features calculated in short intervals, and then probabilistic models are estimated for the feature distributions. Gaussian mixture models and hidden Markov models are tested in this study. The similarity of the signals is measured by the congruence between the feature distributions or by a cross-likelihood ratio test. We calculate the Kullback-Leibler divergence between the distributions by sampling the distributions at the points of the observations vectors. The cross-likelihood ratio test is evaluated using the likelihood of the first signal being generated by the model of the second signal, and vice versa. Simulations were conducted to test the accuracy of the proposed methods on query-by-example of audio. On a database consisting of of speech, music, and environmental sounds the proposed methods enable better retrieval accuracy than the existing methods.

ABSTRACT

No abstract

Calculating the similarity estimates between the query sample and the database samples becomes an exhaustive task with large, usually continuously updated multimedia databases. In this paper, a fast and low complexity transformation from the original feature space into k-dimensional vector space and clustering are proposed to alleviate the problem. First k keysamples are chosen randomly from the database. These samples and a distance function specify the transformation from the series of feature vectors into k-dimensional vector space where database (re)clustering can be done fast with plurality of traditional clustering technique whenever required. In the experiments, similarity between the samples was calculated by using the Euclidean distance between their associated feature vector probability density functions. The k-means algorithm was used to cluster the transformed samples in the vector space. The experiments show that considerable time and computational savings are achieved while there is only a marginal drop in performance.