The SOM Robustness Capacity for Phonemes Recognition in Adverse Environment
Mohamed SalahSalhi, Najet Arous, Noureddine Ellouze · International Journal of Computer Applications · 2012
In this paper, we describe the formatting guidelines for IJCA Journal Submission.The SOM, for kohonen Self Organizing Map, has proven to be a classifier of high caliber in the field of speech recognition signals breasts.Thus, several versions and enhancements were applied on this tool such as recurrent SOM 'RSOM', the growing recurrent SOM 'GRSOM' and the growing hierarchical SOM GHSOM, to consider the integration of sound variability.This paper aims to detect the ability of the SOM robustness in terms of phonemes recognition for continuous speech in a noisy environment.This idea represents, in fact, the real case for speech recognition.