Importance sampling to compute likelihoods of noise-corrupted speech.

Rogier C. van Dalen Mark John Francis Gales

Published in: Comput. Speech Lang. (2013)

Keyphrases

importance sampling
monte carlo
kalman filter
markov chain
particle filter
approximate inference
markov chain monte carlo
particle filtering
rare events
simulated annealing
random variables
three dimensional
hidden markov models
posterior probability