Vision, Speech, and Natural Language are three core areas of Artificial Intelligence. Research in these areas has motivated and influenced research in other areas including multi-processing, parallel processing, robotics, and learning. Speech understanding and speech recognition are two related tasks. Understanding speech means getting the meaning of an utterance such that one can respond properly. Speech recognition is transcribing the speech without necessarily knowing the meaning of the utterance. Automatic speech recognition system has a long history of being difficult problem. Speech recognition is the field of artificial intelligence through which an acoustic waveform is converted into text. Hidden markov model (HMM) is very popular and widely used technique for speech recognition. This paper reviews briefly the HMM technique and highlights the benefits and issues related to it.
Automatic Speech Recognition, Hidden Markov Model, Transcribing, Natural Language Processing, Dynamic Time Wrapping, Cepstral Vectors
