Main Article Content
Abstract
An isolated Spoken word Recognition system "SRIJAN" (srujan) has been developed by Aeronautical Development Establishment (ADE) This speaker dependent and robust, system employs energy based end point detection, Linear Predictive Code (LPC) coefficients derived cepstral coefficients as feature vectors and Dynamic Time Warping (DTW) algorithm. The DTW algorithm offers belter system performance by minimizing the effect of speaking rate variation. The optimum end point pair (start and end) obtained by taking the average of different end point pairs, resulting from the marginal variation of the lower and upper energy thresholds, results in the improvement of the system performance and reduces the computational complexities. Many new techniques such as multiple reference patterns, averaged reference pattern, online and interactive online reference pattern updating, Cepstral Mean Subtraction (CMS), etc. have been implemented to enhance the recognition accuracy and simplify the training required.
Keywords
Article Details
References
- Rabiner, L.R. and Sambur, M.R., "An Algorithm for Determining the Endpoints of lsolated Utterances", Bell Syst. Tech. Journal, Vol.54, February 1975, pp.297-315.
- Hiroaki Sakoe and Seibi Chiba,, "Dynarnic Programming Algorithm Optimization for Spoken Word Recognition", IEEE Trans. Acoust, Speech, and Signal Process, Vol.AS SP-26, February 197 8, pp.43 -49 .
- Atal, B.S., "Effectiveness of Linear Prediction Characteristics of the Speech Wave for Automatic Speaker Identification and Verification", Journal of Acoust. Soc. Amer., Vol.55, June 1974, pp.l304-1312.
- Markel, J.D. and Gray, A.H., "Linear Prediction of Speech", Berlin : Springer-Verlag, 1976.
- Mazel, D.S. and Hayes III, M.H., "Reflections on Levinson's Recursion", ICASSP-88, Vol.3, April 1988, pp.1632-1635.
- Richard A. Haddad and Thomas W. Parsons., "Digital Signal Processing, Theory, Applications and Hardware", 1991, New York, Computer Science Press, Appendix-E, pp.582-587.
- Picone, J. et.al., "Signal Modeling Techniques in Speech Recognition", proceedings ofthe 1993 IEEE Automatic Speech Recognition and Understanding
- Workshop, Vol. 8 l, No.9, pp. I 2 I 5- I 247, September, 1993.
- Suresh Balakrishnama., Final Paper on "Speech Recognition Using Mel Cepstrum, Delta Cepstrum and Delta-Delta Features", Submitted to Department of Electrical and Computer Engineering, Mississippi State University, Mississippi State, Mississippi 39762, to fulfill the requirements for ECE 89993 : Fundamentals of Speech Recognition, December, 5, 1998,
- Martin, T., "Application of limited Vocabulary Recognition Systems", in Rec. 1974 Symp. Speech Recognition, Dr.Reddy, Ed. New York, Academic, 1975, pp.55-71.
References
Rabiner, L.R. and Sambur, M.R., "An Algorithm for Determining the Endpoints of lsolated Utterances", Bell Syst. Tech. Journal, Vol.54, February 1975, pp.297-315.
Hiroaki Sakoe and Seibi Chiba,, "Dynarnic Programming Algorithm Optimization for Spoken Word Recognition", IEEE Trans. Acoust, Speech, and Signal Process, Vol.AS SP-26, February 197 8, pp.43 -49 .
Atal, B.S., "Effectiveness of Linear Prediction Characteristics of the Speech Wave for Automatic Speaker Identification and Verification", Journal of Acoust. Soc. Amer., Vol.55, June 1974, pp.l304-1312.
Markel, J.D. and Gray, A.H., "Linear Prediction of Speech", Berlin : Springer-Verlag, 1976.
Mazel, D.S. and Hayes III, M.H., "Reflections on Levinson's Recursion", ICASSP-88, Vol.3, April 1988, pp.1632-1635.
Richard A. Haddad and Thomas W. Parsons., "Digital Signal Processing, Theory, Applications and Hardware", 1991, New York, Computer Science Press, Appendix-E, pp.582-587.
Picone, J. et.al., "Signal Modeling Techniques in Speech Recognition", proceedings ofthe 1993 IEEE Automatic Speech Recognition and Understanding
Workshop, Vol. 8 l, No.9, pp. I 2 I 5- I 247, September, 1993.
Suresh Balakrishnama., Final Paper on "Speech Recognition Using Mel Cepstrum, Delta Cepstrum and Delta-Delta Features", Submitted to Department of Electrical and Computer Engineering, Mississippi State University, Mississippi State, Mississippi 39762, to fulfill the requirements for ECE 89993 : Fundamentals of Speech Recognition, December, 5, 1998,
Martin, T., "Application of limited Vocabulary Recognition Systems", in Rec. 1974 Symp. Speech Recognition, Dr.Reddy, Ed. New York, Academic, 1975, pp.55-71.