The authors propose a new model parameter compensation algorithm based on parallel model combination (PMC). It differs from PMC in that the amount of adaptation for the parameters is varied depending on the states and mixture components of continuous density HMM. A state-dependent association factor which determines the adaptation is employed and obtained by an EM algorithm
Published in:
Electronics Letters
(Volume:34
,
Issue:
8
)
Date of Publication: 16 Apr 1998