Swaranjali : Isolated Word Recognition for Hindi Language using VQ and HMM

This paper describes the implementation of Swaranjali, an experimental, speaker-dependent, real-time, isolated word recognizer for Hindi. The motivation and the advantages of choosing Hindi as the language for recognition are discussed here. The results obtained with Swaranjali for tests conducted on a vocabulary of Hindi digits for 2 male speakers are presented in the end. The rest of the paper discusses the implementation of the system. The scheme proposed uses a standard implementation, with some modifications to the noise detection/elimination algorithm and the HMM training algorithm. These are also presented in this paper.

[1]  A. Poritz,et al.  Hidden Markov models: a guided tour , 1988, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.

[2]  Biing-Hwang Juang,et al.  Fundamentals of speech recognition , 1993, Prentice Hall signal processing series.

[3]  Robert M. Gray,et al.  An Algorithm for Vector Quantizer Design , 1980, IEEE Trans. Commun..

[4]  M.G. Bellanger,et al.  Digital processing of speech signals , 1980, Proceedings of the IEEE.

[5]  L. R. Rabiner,et al.  On the application of vector quantization and hidden Markov models to speaker-independent, isolated word recognition , 1983, The Bell System Technical Journal.

[6]  Biing-Hwang Juang,et al.  The past, present, and future of speech processing , 1998, IEEE Signal Process. Mag..