Maximum mutual information estimation of HMM parameters for continuous speech recognition using the N-best algorithm.
Yen-Lu ChowPublished in: ICASSP (1990)
Keyphrases
- estimation algorithm
- expectation maximization
- mutual information
- maximum likelihood estimation
- optimization algorithm
- k means
- dynamic programming
- optimal solution
- computational complexity
- search space
- em algorithm
- optimal parameters
- neural network
- computational cost
- np hard
- preprocessing
- forward backward
- matching algorithm
- convergence rate
- recognition algorithm
- parameter settings
- image processing
- parameter tuning
- parameters estimation
- information theoretic criterion
- least squares
- parameter space
- maximum likelihood
- probabilistic model
- hidden markov models
- cost function
- significant improvement
- image segmentation
- computer vision
- machine learning