A study on HMM based speech recognition system

Speech is the best, old and natural mode of communication among the human being without using any tool. It is the easiest way to express human feelings. So if human speech could be used to communicate with computer then it is quiet helpful for human beings to interact with computer without using keyboard interface. Hence the most important goal of speech recognition is to enable human being to communicate with the computer more naturally and effectively by means of using some algorithm. This system changed the way of how to communicate with machine. By using this system we want to make the computer to be able to understand the spoken words as speech plays an important role for the interaction with the computer. This paper gives a basic overview of HMM based speech recognition system, how does it actually work. HMM stands for Hidden Markov Model which is basically a statistical model. The speech recognition system that we discuss here is based on this statistical model.