P.S. Gopalakrishnan, Dimitri Kanevsky, et al.
IEEE Trans. Inf. Theory
Paper
Estimation of Probabilities in the Language Model of the IBM Speech Recognition System
Abstract
The language model probabilities are estimated by an empirical Bayes approach in which a prior distribution for the unknown probabilities is itself estimated through a novel choice of data. The predictive power of the model thus fitted is compared by means of its experimental perplexity [1] to the model as fitted by the Jelinek-Mercer deleted estimator and as fitted by the Turing-Good formulas for probabilities of unseen or rarely seen events. Copyright © 1984 by The Institute of Electrical and Electronics Engineers, Inc.
Related
Arthur Nádas
IEEE Transactions on Acoustics, Speech, and Signal Processing
Conference paper
On the game of googol as the secretary problem
Stephen Silverman, Arthur Nádas
ISIT 1991
Toby G. Rossman, Ekaterina I. Goncharova, et al.
Mutation Research - Fundamental and Molecular Mechanisms of Mutagenesis