P. Bergl and R. Cmejla (Czech Republic)
Signal processing; HMM; Divergence measure; Speech database.
An algorithm for improvement of boundaries estimated via HMM is suggested, using numerous divergence mea sures (spectral distance, cepstral distance, difference of the cepstral coefficients, Kullback-Leibler divergence, distance given by the General Likelihood Ratio, distance defined by the family of Recursive Bayesian Changepoint Detec tors, the Bhattacharyya divergence, the Mahalanobis mea sure, the L2 metric and the Jeffreys-Matusita measure). A method for choosing of appropriate measure for each boundary is stated, and is demonstrated on the training set of signals. The ability to improve boundary positions is verified on the testing set, and it is found that a combina tion of methods brings better results than using only one measure.
Important Links:
Go Back
