Title :
A new algorithm to design optimal Hidden Markov Models
Author :
Dologlou, I. ; Bakamidis, S. ; Carayannis, G.
Author_Institution :
Institute for Language and Speech Processing (ILSP) Artemidos and Epidavrou Str., Paradeisos Amaroussiou, Athens 151 25, Greece
Abstract :
A new algorithm for the design of Hidden Markov Models (HMM) from observed symbol and bisymbol probabilities is presented. The algorithm provides a global optimum and makes use of linear vectorial models of sequences of probabilistic vectors estimated during an off-line learning process. Moreover, a method to enhance observed data so as to comply with the constraints of HMM strings is proposed. An optimal estimate of the number of states of the HMM for the given observed data is also provided.
Conference_Titel :
Signal Processing Conference, 2000 10th European
Print_ISBN :
978-952-1504-43-3
