A Bayesian model selection criterion for HMM topology optimization

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Volumn 1, Issue , 2002, Pages 989-992

  Biem, Alain ^a   Ha, Jin Young ^b   Subrahmonia, Jayashree ^a  

^a IBM T J WATSON RESEARCH CENTER   (United States)

^b Kangwon National University   (South Korea)

Author keywords

[No Author keywords available]

Indexed keywords

BAYESIAN INFORMATION CRITERION; HANDWRITING RECOGNITION TASK; HIDDEN MARKOV MODEL;

ALGORITHMS; LAPLACE TRANSFORMS; MARKOV PROCESSES; MATHEMATICAL MODELS; OPTIMIZATION; PARAMETER ESTIMATION;

SPEECH RECOGNITION;

EID: 0036289982     PISSN: 15206149     EISSN: None     Source Type: Journal    
DOI: 10.1109/ICASSP.2002.5743960     Document Type: Article

References (8)

1. Maximum likelihood successive state splitting
2. Hidden markov model induction by Bayesian model merging
3. Estimating the dimension of a model
4. HMM topology optimization for handwriting recognition
5. Bayesian interpolation
6. Bayes factors
7. MML and Bayesianism: Similarities and differences (introduction to minimum encoding inference - Part II
8. EDF statistics for goodness of fit and some comparisons