Zobrazeno 1 - 10
of 130
pro vyhledávání: '"K. Visweswariah"'
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
Publikováno v:
IEEE Transactions on Information Theory. 48:1061-1081
The Burrows Wheeler transform (1994) is a reversible sequence transformation used in a variety of practical lossless source-coding algorithms. In each, the BWT is followed by a lossless source code that attempts to exploit the natural ordering of the
Publikováno v:
IEEE Transactions on Information Theory. 47:1461-1472
A universal variable-to-fixed length algorithm for binary memoryless sources which converges to the entropy of the source at the optimal rate is known. We study the problem of universal variable-to-fixed length coding for the class of Markov sources
Publikováno v:
IEEE Transactions on Information Theory. 46:1633-1637
We investigate the performance of the Lempel-Ziv (1978) incremental parsing scheme on nonstationary sources. We show that it achieves the best rate achievable by a finite-state block coder for the nonstationary source. We also show a similar result f
Publikováno v:
IEEE Transactions on Information Theory. 46:2237-2241
We consider the problem of determining when a given source can be used to approximate the output due to any input to a given channel. We provide achievability and converse results for a general source and channel. For the special case of a full-rank
Publikováno v:
IEEE Transactions on Information Theory. 44:462-471
A random number generator generates fair coin flips by processing deterministically an arbitrary source of nonideal randomness. An optimal random number generator generates asymptotically fair coin flips from a stationary ergodic source at a rate of
Publikováno v:
ICME
Multi-stream hidden Markov models (HMMs) have recently been very successful in audio-visual speech recognition, where the audio and visual streams are fused at the final decision level. In this paper we investigate fast feature space speaker adaptati
Autor:
Sanjeev R. Kulkarni, K. Visweswariah
Publikováno v:
Proceedings. 1998 IEEE International Symposium on Information Theory (Cat. No.98CH36252).
A universal variable-to-fixed length code for the class of Markov sources with finite output alphabets is considered. The performance of the code for large dictionary sizes is investigated and it is shown that the code is optimal in the sense that no
Publikováno v:
2000 IEEE International Symposium on Information Theory (Cat. No.00CH37060).
The Burrows-Wheeler transform is a block-sorting algorithm which has been shown empirically to be useful in compressing text data. In this paper we study the output distribution of the transform for i.i.d. sources, tree sources and stationary ergodic
Publikováno v:
6th International Conference on Spoken Language Processing (ICSLP 2000).