Spatio-Temporal and Complex-Valued Models based on SOM map applied to Speech Recognition
Résumé
Speech perception and recognition using biologically inspired models is a challenging issue not well explored yet. The paper presents two spatiotemporal biologically inspired methods for the preprocessing and learning of speech signal based on SOM (Self Organization Map). The experimental results are very encouraging and provide a framework which could be more studied to overcome present limitation of speech perception and recognition technology.