Zobrazeno 1 - 10
of 62
pro vyhledávání: '"Ramesh A. Gopinath"'
Publikováno v:
IEEE Transactions on Speech and Audio Processing. 13:1144-1160
A standard approach to automatic speech recognition uses hidden Markov models whose state dependent distributions are Gaussian mixture models. Each Gaussian can be viewed as an exponential model whose features are linear and quadratic monomials in th
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::608d76850770614741533f3227282309
https://doi.org/10.1109/tsa.2005.851965
https://doi.org/10.1109/tsa.2005.851965
Autor:
Ramesh A. Gopinath
Publikováno v:
IEEE Transactions on Signal Processing. 53:1794-1806
Phaselets are a set of dyadic wavelets that are related in a particular way such that the associated redundant wavelet transform is nearly shift-invariant. Framelets are a set of functions that generalize the notion of a single dyadic wavelet in the
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::1de8adbb7ecad12f13c08834b1f0e1d6
https://doi.org/10.1109/tsp.2005.845471
https://doi.org/10.1109/tsp.2005.845471
Autor:
Peder A. Olsen, Ramesh A. Gopinath
Publikováno v:
IEEE Transactions on Speech and Audio Processing. 12:37-46
This paper proposes a new covariance modeling technique for Gaussian mixture models. Specifically the inverse covariance (precision) matrix of each Gaussian is expanded in a rank-1 basis i.e., /spl Sigma//sub j//sup -1/=P/sub j/=/spl Sigma//sub k=1//
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e5f7ffb12ebbfac4c1913fe403c6eba7
https://doi.org/10.1109/tsa.2003.819943
https://doi.org/10.1109/tsa.2003.819943
Autor:
Ramesh A. Gopinath
Publikováno v:
IEEE Transactions on Signal Processing. 51:1792-1805
This paper introduces an approximately shift invariant redundant dyadic wavelet transform - the phaselet transform - that includes the popular dual-tree complex wavelet transform of Kingsbury (see Phil. R. Soc. London A, Sept. 1999) as a special case
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::6e996927622a5c4262cf7f1e523ba62d
https://doi.org/10.1109/tsp.2003.812833
https://doi.org/10.1109/tsp.2003.812833
Autor:
Ramesh A. Gopinath
Publikováno v:
IEEE Transactions on Signal Processing. 51:182-192
This paper describes the design of finite impulse response (FIR) delay filters that minimize a squared error and have prescribed number of zeros at /spl omega/=/spl pi/ and prescribed magnitude and group delay flatness at /spl omega/=0. An important
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::8b52bba97aa18d5b88bf2439a4d0ec05
https://doi.org/10.1109/tsp.2002.806589
https://doi.org/10.1109/tsp.2002.806589
Autor:
Sabine Deligne, Ramesh A. Gopinath, Peder A. Olsen, Benoît Maison, Harry Printz, Satyanarayana Dharanipragada
Publikováno v:
IEEE Transactions on Speech and Audio Processing. 10:551-561
This paper describes a robust, accurate, efficient, low-resource, medium-vocabulary, grammar-based speech recognition system using hidden Markov models for mobile applications. Among the issues and techniques we explore are improving robustness and e
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e8a05904d7de89fe808eefd638e7d7ca
https://doi.org/10.1109/tsa.2002.804541
https://doi.org/10.1109/tsa.2002.804541
Autor:
D. Kanvesky, M. J. F. Gales, Ellen Eide, Scott Shaobing Chen, Ramesh A. Gopinath, Peder A. Olsen
Publikováno v:
Speech Communication. 37:69-87
This paper describes the IBM approach to Broadcast News (BN) transcription. Typical problems in the BN transcription task are segmentation, clustering, acoustic modeling, language modeling and acoustic model adaptation. This paper presents new algori
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::9b43abb5a16e34039df0a1fae462d470
https://doi.org/10.1016/s0167-6393(01)00060-7
https://doi.org/10.1016/s0167-6393(01)00060-7
Autor:
Ramesh A. Gopinath
Publikováno v:
IEEE Transactions on Signal Processing. 44:984-987
Recently, a simple and versatile method for the design of linear phaser FIR filters with spline transition bands and optimal in a least-squared sense was introduced. The following question is raised: Given an arbitrary window, say, for example, a Ham
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::21eb54a1a84babc405ca750402e56557
https://doi.org/10.1109/78.492551
https://doi.org/10.1109/78.492551
Autor:
Ramesh A. Gopinath, C.S. Burrus
Publikováno v:
IEEE Transactions on Signal Processing. 43:666-680
A complete factorization of all optimal (in terms of quick transition) time-varying FIR unitary filter bank tree topologies is obtained. This has applications in adaptive subband coding, tiling of the time-frequency plane and the construction of orth
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e35a061676779fc51c7762b6adfa2a38
https://doi.org/10.1109/78.370621
https://doi.org/10.1109/78.370621
Autor:
C.S. Burrus, Ramesh A. Gopinath
Publikováno v:
IEEE Transactions on Signal Processing. 42:812-824
Solutions to the problem of design of rational sampling rate filter banks in one dimension has previosly been proposed. The ability to interchange the operations of upsampling, downsampling, and filtering plays an important role in these solutions. T
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::9ba41c60fde97dc32d11c2cae136f3a9
https://doi.org/10.1109/78.285645
https://doi.org/10.1109/78.285645