Title :
Statistically reliable deleted interpolation
Author :
Kim, Nam Soo ; Un, Chong Kwan
Author_Institution :
Speech Technol. Lab., Samsung Adv. Inst. of Technol., Kyungki-Do, South Korea
fDate :
5/1/1997 12:00:00 AM
Abstract :
The output probability distributions (PDs) in each state of a discrete HMM suffer from sparseness, causing inaccurate modeling of probabilistic characteristics of speech features within the state. A desirable solution to the problem arising from insufficient training data is to interpolate a maximum likelihood (ML) estimate of a PD with some other estimates that are, to some extent, able to strengthen the robustness of the PD with respect to unseen data. We propose a statistically reliable deleted interpolation (DI) approach. The DI is an efficient technique for interpolating several probability distribution (PD) estimates, and usually different interpolating weights are used for each predetermined range of PD counts. Our approach attempts to piecewise linearly approximate the interpolating weight curve based on some reasoning concerned with statistical reliability of sample-based estimates
Keywords :
hidden Markov models; interpolation; maximum likelihood estimation; piecewise-linear techniques; probability; signal sampling; speech processing; statistical analysis; discrete HMM; interpolating weight curve; interpolating weights; maximum likelihood estimate; output probability distributions; piecewise linear approximation; probabilistic characteristics; sample based estimates; speech features modeling; statistically reliable deleted interpolation; Degradation; Hidden Markov models; Interpolation; Maximum likelihood estimation; Parameter estimation; Probability distribution; Robustness; Speech recognition; Training data; Yield estimation;
Journal_Title :
Speech and Audio Processing, IEEE Transactions on
