Feature Extraction and Modeling Techniques in Speech Recognition: A Review

Abstract

Major advancement is being noted frequently on both the exploitation and technology of spoken language systems and automatic speech recognition (ASR). Speech recognition is the procedure of automatically identifying the pronounced words of speaker based on information in voice signal. In this paper our main objectives is to review different feature extraction techniques used in Speech recognition and then several techniques for modeling the speech signal. Furthermore we also discuss the matching and prediction process. Finally this paper also provides some performance measure for speech recognition techniques.