Introduction

Abstract

Human preference for speech as a natural form of communication has increased the interest of many research groups over the years in developing robust Automatic Speech Recognition (ASR) systems. These systems allow a natural operation of several devices such as mobile phones, navigations systems and language translators. Additionally, novice users would be able to get easily familiar with a particular device if the interface is based on speech. Moreover, there are more advanced tasks such as call-centers where a fully interaction between humans and machines is required. In these tasks, a Spoken Dialog System is needed [Minker 04] where the ASR module plays an important role in the robustness of the system.