Title :
Influence of initialisation and stop criteria on HMM based recognisers
Author :
Ferrer, M.A. ; Alonso, I.G. ; Travieso, C.M.
Author_Institution :
Dept. de Senales y Comunicaciones, Univ. de las Palmas de Grun Canaria, Spain
fDate :
6/22/2000 12:00:00 AM
Abstract :
A study is presented into the importance of two commonly overlooked factors influencing generalisation ability in the field of hidden Markov model (HMM) based recogniser training algorithms by means of a comparative study of four initialisation methods and three stop criteria in different applications. The results show that better results have been found with the equal-occupancy initialisation method and the fixed-threshold stop criterion
Keywords :
generalisation (artificial intelligence); hidden Markov models; learning (artificial intelligence); speech recognition; HMM based recognisers; equal-occupancy initialisation method; fixed-threshold stop criterion; generalisation ability; initialisation; speech recognition; training algorithms;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:20000826
