Voice Based user Machine Interface for Punjabi using Hidden Markov Model

With new advancements in the field of digital signal processing has increase the use of speech in different areas like speech compression, synthesis, enhancement and recognition. In this paper the concept of Hidden Markov Model has been used for speech command recognition (speaker dependent) and turned a machine into intelligent machine which will respond to human speech and activate the operating system commands. Hidden Markov Model implements speech recognizer in three steps. The first step is preprocessing that includes endpoint detection, framing and windowing of the speech signal. The second step includes speech feature extraction using Linear Predictive Coefficients, Cepstral analysis and then Vector quantization of speech features is done, the last step involves the Hidden Markov Model based classifier for speech command recognition.