Abstract

In this paper the implementation of the word level speech recognition system for Punjabi language is explained because it is a highly prosodic language. Here HTK Toolkit along with Julius Toolkit is used. First step is data collection and two hours data is collected in read speech mode. Second step is data preparation, in which hmmlist, grammar and dictionary files are created. Once the data is prepared, 75% and 25% of data is used for training and testing respectively. The experimental results show that the accuracy of the system comes out to be 57.54%

References

• HTK “Hidden Markov Model Toolkit”, available at “http://htk.eng.cam.ac.uk”, 2012.
• L.R. Rabiner , “A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition”, Proc. of the IEEE Vol. 77, Issue 2,pp. 257–286,1989.
• “Introduction of Julius,” http://julius.osdn.jp/en_index.php
• Azmi, M., Tolba, H., Mahdy, S., and Fashal, M. (2008). "Syllable-based automatic Arabic speech recognition", Proceedings of the 7th WSEAS International Conference on Signal Processing, Robotics and Automation, World Scientific and Engineering Academy and Society, WSEAS, 246-250.
• R. Kumar “Comparison of HMM and DTW for Isolated Word Recognition of Punjabi Language” In Proceedings of Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications, Sao Paulo, Brazil. Vol. 6419 of Lecture Notes in Computer Science (LNCS), pp. 244– 252, Springer Verlag, November 8-11, 2010.
• R. Kumar and M. Singh, “Spoken isolated Word Recognition of Punjabi Language Using dynamic time Warping Technique” Demo in Proceedings of Information System for Indian Languages, Punjabi University, Patiala, India, March 9 - 11, 2011. Vol. 139 of Communication in Computer and Information Science (CCIS), Page 301, Springer Verlag.
• K. Kumar, R. K. Aggarwal, and A. Jain “A Hindi speech recognition system for connected words using HTK” International Journal Computational Systems Engineering, Vol. 1, No. 1, 2012
• B. A. Q. Al-Qatab and R. N. Ainon, “Arabic Speech Recognition Using Hidden Markov Model Toolkit (HTK)”, Paper presented at International Symposium in Information Technology (ITSim). Kuala Lumpur, June 15-17, 2010.
• Mandal, S., Das, D., Mitra, P.: SHRUTI-II: A Verncular Speech Recognition System in Bengali and an Application.
• S. Young, “Hidden Markov Model Toolkit: Design and Philosophy,” CUED/F-INENG/TR.152, Cambridge University Engineering Department, Sept. 1994.
• Lee, K. F., Hon, H. W., Hwang, M. Y., and Mahajan, S. (1989), "The SPHINX speech recognition system", Proceedings of the IEEE International Conference in Acoustics, Speech and Signal Processing.
• Woodland, P. C., Leggetter, C. J., Odell, J. J., Valtchev, V., and Young, S. J. (1995). "The HTK large vocabulary speech recognition system", IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP, 1, 73-76.
• Choudhary, A., Chauhan, M. R., and Gupta, M. G. (2013). "Automatic Speech Recognition System for Isolated and Connected Words of Hindi Language By Using Hidden Markov Model Toolkit (HTK)".
• Lee, K. F., and Hon, H. W. (1989). "Speaker-independent phone recognition using hidden Markov models", IEEE Transactions on Acoustics, Speech and Signal Processing, 37(11), 1641-1648.
• Steve Young, Gunnar Evermann, Mark Gales. HTK Book (for HTK version 3.4). England, Cambridge University of Engineering Department, 2006.
• Ming, J., and Smith, F. J. (1998). "Improved phone recognition using Bayesian tri phone models", Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, 1, 409-412