Speeding Up HMM Decoding and Training by Exploiting Sequence Repetitions.
Yury LifshitsShay MozesOren WeimannMichal Ziv-UkelsonPublished in: Algorithmica (2009)
Keyphrases
- hidden markov models
- viterbi algorithm
- baum welch
- sequential data
- sequence classification
- forward backward
- training phase
- decoding process
- training set
- online learning
- conditional random fields
- speech recognition
- training examples
- discriminative training
- dynamic programming
- handwritten word recognition
- training process
- training algorithm
- markov model
- sequence analysis
- training stage
- training samples
- multi stream
- supervised learning
- minimum classification error