An Overview of Speech Recognition Using HMM?

The Speech is most prominent & one of the natural forms of communication among of human being. The speech is a signal of infinite information. There are different aspects related to speech like speech recognition, speech verification, speech synthesis, speaker recognition, speaker identification etc. The purpose of this project is to study a speech recognition system using HMM. The goal of speech recognition is to determine which speech is present based on spoken information. The system uses MFCC for feature extraction and HMM for pattern training. The success of MFCC combined with their robust and costeffective computation, turned them into a standard choice in speech recognition applications. And HMM provides a highly reliable way of recognizing speech.