A modified Baum-Welch algorithm for hidden Markov models with multiple observation spaces

IEEE Transactions on Speech and Audio Processing

Volumn 9, Issue 4, 2001, Pages 411-416

  Baggenstoss, Paul M ^a,b  

^a IEEE   (United States)

^b NAVAL UNDERSEA WARFARE CENTER   (United States)

Author keywords

Baum Welch algorithm; Class specific; EM algorithm; Expectation maximization; Gaussian mixtures; Hidden Markov model (HMM); Parameter estimation; Sufficient statistics

Indexed keywords

BAUM-WELCH ALGORITHMS; HIDDEN MARKOV MODELS (HMM);

ALGORITHMS; MARKOV PROCESSES; MAXIMUM LIKELIHOOD ESTIMATION; PARAMETER ESTIMATION; PROBABILITY DENSITY FUNCTION;

SPEECH RECOGNITION;

EID: 0035340676     PISSN: 10636676     EISSN: None     Source Type: Journal    
DOI: 10.1109/89.917686     Document Type: Article

References (11)

1. Class-specific features in classification
2. Sufficiency, classification, and the class-specific feature theorem
3. A tutorial on hidden Markov models and selected applications in speech recognition
4. A maximization technique occurring in the statistical analysis of probabilistic functions of markov chains
5. Maximum likelihood estimation for mixture multivariate stochastic observations of Markov chains
6. Accurate evaluation of multidimensional probability density functions (pdfs) for signal processing
7. The joint distribution of serial correlation coefficients