Tied Mixture Continuous Parameter Modelling for Speech Recognition
Original Publication Date: 1989-Dec-01
Included in the Prior Art Database: 2005-Jan-29
Discrete and continuous parameter approaches to the acoustic-modelling problem in automatic speech recognition are unified through a class of general hidden Markov models, whose output probability distributions are specified using tied mixtures of simple multivariate densities. Speech recognition experiments performed on large vocabulary office correspondence tasks demonstrate some of the resulting benefits.