Abstract : This paper deals with two new algorithms for Multi-Band Speech Recognition : the first one enhances the robustness of the system to stationary noise, and the second one is a global training algorithm which allows the system to reach its global optimum. Experimental results confirm the good behavior of these two methods.
https://hal.inria.fr/inria-00098917
Contributeur : Publications Loria
<>
Soumis le : mardi 26 septembre 2006 - 08:40:17
Dernière modification le : vendredi 9 février 2018 - 13:20:01