next up previous contents
Next: Task 5.2: Status Up: Task 5.2: Global Discrimination Previous: Task 5.2: Global Discrimination

Task 5.2: Task Objective

The initial goal of this task was to investigate, develop and test further the new hybrid HMM/ANN system initiated in WERNICKE. This new technique, referred to as REMAP (Recursive estimation and Maximization of a Posteriori Probabilities), allows full discriminant training of HMM/ANN systems, leading to globally discriminant models [Bourlard et al., 1995].

Christophe Ris