Speech Recognition using Hidden Markov Models in Embedded Platform

Speech recognition used widely in environment of mobile application. The Study of speech recognition is one of the main topics of journal in artificial intelligence which doesn’t take out meaningful result. The reason of such result is to relate with difficulty of feature extraction. Liner Predictive Coding, Hidden Markov Model, Artificial Neural Network are known to be effective in the same way as for the voice signal processing. Mel frequency cepstral coefficients are the most popular method for extracting speech features from the speech recognition field. This paper proposes a method for recognizing speech using Mel-frequency information. In this paper we propose Automatic Speech Recognition (ASR) technique using Mel-Frequency Cepstral Coefficient extraction and Hidden Markov Model in mobile environment. Our method is used to speech enhancement technique. We’ll try to implement ASR system in mobile environment using embedded platform.