An open source approach for continuous speech recognition using hidden Markov models

Abstract: Speech is a basic mode of communication between us and most natural efficient form of exchanging information. Speech Recognition is a conversion of an acoustic waveform to text. Speech can be isolated, connected and continuous type. The goal of this work is to recognize a Continuous Speech using Mel Frequency Cepstrum Coefficients (MFCC) to extract the features of Speech signal, Hidden Markov Models (HMM) for pattern recognition and Viterbi Decoder for decoding of speech signal. Continuous Speech files of the TIMIT standard database are used for the work. The recognition success rate is calculated for the entire database, separate Training and Testing files are found in the database and we also prepared a small set of database used in our work. For the complete process we used Hidden Markov Model Tool Kit (HTK) which is an Open source tool developed by Cambridge University Engineering Department (CUED), which contains a set of standard C Programs for feature extraction, model building and for decoding purposes, for the entire work Linux Operating System fedora is used, The objective of the work is to develop an open source HTK based Continuous Speech Recognition & to obtain better recognition accuracy for large vocabulary size.

Keywords: Speech recognition, feature extraction, pattern recognition, HTK, TIMIT, HMM