Zobrazeno 1 - 10
of 61
pro vyhledávání: '"S. Dharanipragada"'
Publikováno v:
IEEE Transactions on Audio, Speech and Language Processing. 15:224-234
This paper describes a robust feature extraction technique for continuous speech recognition. Central to the technique is the minimum variance distortionless response (MVDR) method of spectrum estimation. We consider incorporating perceptual informat
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::4c71dbff770120e4e7c45bd01bbaf0f6
https://doi.org/10.1109/tasl.2006.876776
https://doi.org/10.1109/tasl.2006.876776
Autor:
S. Dharanipragada, K. Visweswariah
Publikováno v:
IEEE Transactions on Audio, Speech and Language Processing. 14:1255-1266
We introduce a class of Gaussian mixture models (GMMs) in which the covariances or the precisions (inverse covariances) are restricted to lie in subspaces spanned by rank-one symmetric matrices. The rank-one basis are shared between the Gaussians acc
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::c633a91a6dda46962cf156bd4f4ea086
https://doi.org/10.1109/tsa.2005.860835
https://doi.org/10.1109/tsa.2005.860835
Autor:
M. Padmanabhan, S. Dharanipragada
Publikováno v:
IEEE Transactions on Speech and Audio Processing. 13:512-519
In this paper, we consider the problem of quantifying the amount of information contained in a set of features, to discriminate between various classes. We explore these ideas in the context of a speech recognition system, where an important classifi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::c7f2f6c7eb3606f7dce8c05631157546
https://doi.org/10.1109/tsa.2005.848876
https://doi.org/10.1109/tsa.2005.848876
Autor:
S. Dharanipragada, K.S. Arun
Publikováno v:
IEEE Transactions on Signal Processing. 45:2951-2966
The problem of extrapolating discrete-index bandlimited signals from a finite number of samples is addressed in this paper. The algorithm presented in this paper exploits the fact that the set of bandlimited signals that are also essentially time-lim
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::7e5ea6affcaa3292a6602f11bdebf4a1
https://doi.org/10.1109/78.650256
https://doi.org/10.1109/78.650256
Autor:
K.S. Arun, S. Dharanipragada
Publikováno v:
IEEE Transactions on Signal Processing. 44:546-561
Resolution analysis for the problem of signal recovery from finitely many linear measurements is the subject of this paper. The classical Rayleigh limit serves only as a lower bound on resolution since it does not assume any recovery strategy and is
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::cdd692df64feff1a2cfea26e4bf1c738
https://doi.org/10.1109/78.489028
https://doi.org/10.1109/78.489028
Autor:
K.S. Arun, S. Dharanipragada
Publikováno v:
IEEE Transactions on Signal Processing. 44:248-266
This paper addresses the problem of recovering a signal that is constrained to lie in a convex set, from linear measurements. The current standard is the alternating projections paradigm (POCS), which has only first-order convergence in general. We p
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::0eff7ff8b942ff33442370dd0f239cff
https://doi.org/10.1109/78.485922
https://doi.org/10.1109/78.485922
Autor:
K.S. Arun, S. Dharanipragada
Publikováno v:
1990 Conference Record Twenty-Fourth Asilomar Conference on Signals, Systems and Computers, 1990..
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::4c30a7d46dbd878ae1ef5d50de2a45d2
https://doi.org/10.1109/acssc.1990.523478
https://doi.org/10.1109/acssc.1990.523478
Autor:
K.S. Arun, S. Dharanipragada
Publikováno v:
The Digital Signal Processing workshop.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::0d945eec4e855fd61aa6a76d46d6d546
https://doi.org/10.1109/dspws.1992.665576
https://doi.org/10.1109/dspws.1992.665576
Autor:
Umit H. Yapanel, S. Dharanipragada
Publikováno v:
ICASSP (1)
This paper describes a robust feature extraction technique for continuous speech recognition. Central to the technique is the minimum variance distortionless response (MVDR) method of spectrum estimation. We incorporate perceptual information directl
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ca1837724787997963ef89ad519872a8
https://doi.org/10.1109/icassp.2003.1198863
https://doi.org/10.1109/icassp.2003.1198863
Autor:
S. Dharanipragada
Publikováno v:
2002 IEEE International Symposium on Circuits and Systems. Proceedings (Cat. No.02CH37353).
We present new feature extraction and feature transformation techniques that significantly improve the robustness of continuous speech recognition. We report results on two tasks - an in-car embedded speech recognition task and a speaker-phone based
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::c3f986033825e3dc2724165322a4e6dd
https://doi.org/10.1109/iscas.2002.1011488
https://doi.org/10.1109/iscas.2002.1011488