Results 1  10
of
134
Signal modeling techniques in speech recognition
 PROCEEDINGS OF THE IEEE
, 1993
"... We have seen three important trends develop in the last five years in speech recognition. First, heterogeneous parameter sets that mix absolute spectral information with dynamic, or timederivative, spectral information, have become common. Second, similariry transform techniques, often used to norm ..."
Abstract

Cited by 180 (5 self)
 Add to MetaCart
We have seen three important trends develop in the last five years in speech recognition. First, heterogeneous parameter sets that mix absolute spectral information with dynamic, or timederivative, spectral information, have become common. Second, similariry transform techniques, often used to normalize and decorrelate parameters in some computationally inexpensive way, have become popular. Third, the signal parameter estimation problem has merged with the speech recognition process so that more sophisticated statistical models of the signal’s spectrum can be estimated in a closedloop manner. In this paper, we review the signal processing components of these algorithms. These algorithms are presented as part of a unified view of the signal parameterization problem in which there are three major tasks: measurement, transformation, and statistical modeling. This paper is by no means a comprehensive survey of all possible techniques of signal modeling in speech recognition. There are far too many algorithms in use today to make an exhaustive survey feasible (and cohesive). Instead, this paper is meant to serve as a tutorial on signal processing in stateoftheart speech recognition systems and to review those techniques most commonly used. In keeping with this goal, a complete mathematical description of each algorithm has been included in the paper.
Mobile Robot Localization Using Landmarks
, 1997
"... We describe an efficient method for localizing a mobile robot in an environment with landmarks. We assume that the robot can identify these landmarks and measure their bearings relative to each other. Given such noisy input, the algorithm estimates the robot's position and orientation with resp ..."
Abstract

Cited by 155 (4 self)
 Add to MetaCart
(Show Context)
We describe an efficient method for localizing a mobile robot in an environment with landmarks. We assume that the robot can identify these landmarks and measure their bearings relative to each other. Given such noisy input, the algorithm estimates the robot's position and orientation with respect to the map of the environment. The algorithm makes efficient use of our representation of the landmarks by complex numbers. The algorithm runs in time linear in the number of landmarks. We present results of simulations and propose how to use our method for robot navigation.
Entanglement measures and purification procedures
 Physical Review A
, 1998
"... We improve previously proposed conditions each measure of entanglement has to satisfy. We present a class of entanglement measures that satisfy these conditions and show that the Quantum Relative Entropy and Bures Metric generate two measures of this class. We calculate the measures of entanglement ..."
Abstract

Cited by 33 (1 self)
 Add to MetaCart
(Show Context)
We improve previously proposed conditions each measure of entanglement has to satisfy. We present a class of entanglement measures that satisfy these conditions and show that the Quantum Relative Entropy and Bures Metric generate two measures of this class. We calculate the measures of entanglement for a number of mixed two spin 1/2 systems using the Quantum Relative Entropy, and provide an efficient numerical method to obtain the measures of entanglement in this case. In addition, we prove a number of properties of our entanglement measure which have important physical implications. We briefly explain the statistical basis of our measure of entanglement in the case of the Quantum Relative Entropy. We then argue that our entanglement measure determines an upper bound to the number of singlets that can be obtained by any purification procedure. PACS: 03.65.Bz I.
Speaker Normalization with AllPass Transforms
, 1998
"... Speaker normalization is a process in which the shorttime features of speech from a given speaker are transformed so as to better match some speaker independent model. Vocal tract length normalization (VTLN) is a popular speaker normalization scheme wherein the frequency axis of the shorttime spec ..."
Abstract

Cited by 24 (6 self)
 Add to MetaCart
Speaker normalization is a process in which the shorttime features of speech from a given speaker are transformed so as to better match some speaker independent model. Vocal tract length normalization (VTLN) is a popular speaker normalization scheme wherein the frequency axis of the shorttime spectrum associated with a particular speaker's speech is rescaled or warped prior to the extraction of cepstral features. In this work, we develop a novel speaker normalization scheme by exploiting the fact that frequency domain transformations similar to that inherent in VTLN can be accomplished entirely in the cepstral domain through the use of conformal maps. We propose a class of such maps, designated allpass transforms for reasons given hereafter, and rigorously investigate their properties. Theoretical results are provided relating to the transformation of cepstral sequences under these maps. Additionally, all relations necessary to determine maximum likelihood estimates of the mapping parameters are derived for both speaker normalization and adaptation. 1 2 Speaker Normalization with AllPass Transforms 1
Signal synthesis in the presence of an inconsistent set of constraints
 IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS
, 1985
"... In this paper, we present a novel technique for signal synthesis in the presence of an inconsistent set of constraints. This technique represents a general, minimum norm, solution to the class of synthesis problems in which: the desired signal may be characterized as being an element of some Hilbert ..."
Abstract

Cited by 17 (0 self)
 Add to MetaCart
In this paper, we present a novel technique for signal synthesis in the presence of an inconsistent set of constraints. This technique represents a general, minimum norm, solution to the class of synthesis problems in which: the desired signal may be characterized as being an element of some Hilbert Space; each of the N design constraints generates a closed convex set in that space; and those N convex sets generate, or may be resolved into, two disjoint closed convex sets, such that at least one of the two sets is bounded. The synthesis technique employs alternating nearest point maps onto closed convex subsets of a Hilbert Space, and may be viewed as an extension of D. Youla’s “Method of Convex Projections” which addresses the case in which the N closed convex sets, corresponding to the design constraints, possess a nonempty intersection. Section I provides a general introduction to the synthesis problem and to its solution. Section II contains the mathematical justification for the solution technique, while Section III presents an example of the synthesis of a data window for spectral estimation. In Section Iv, we discuss potential extensions of this technique within the area of signal synthesis, as well as to the more general class of constrained optimization problems.
Dispersion of Homogeneous and Inhomogeneous Waves in the Yee FiniteDifference TimeDomain Grid
, 2001
"... The numerical dispersion relation governing the propagation of homogeneous plane waves in a finitedifference timedomain (FDTD) grid is well known. However, homogeneous plane waves, by themselves, do not form a complete basis set capable of representing all valid field distributions. A complete bas ..."
Abstract

Cited by 16 (6 self)
 Add to MetaCart
The numerical dispersion relation governing the propagation of homogeneous plane waves in a finitedifference timedomain (FDTD) grid is well known. However, homogeneous plane waves, by themselves, do not form a complete basis set capable of representing all valid field distributions. A complete basis set is obtained by including inhomogeneous waves where, in the physical world, constant phase planes must be orthogonal to constant amplitude planes for lossless media. In this paper we present a dispersion analysis for both homogeneous and inhomogeneous plane waves in the Yee FDTD grid. We show that, in general, the constant amplitude and constant phase planes of inhomogeneous plane waves are not orthogonal, but they approach orthogonality for fine discretization. The dispersion analysis also shows that, for very coarsely resolved fields, homogeneous waves will experience exponential decay as they propagate and they may propagate faster than the speed of light. Bounds are established for...
Asymptotics of RNA shapes
"... RNA shapes, introduced by Giegerich et al. (17), provide a useful classification of the branching complexity for RNA secondary structures. In this paper, we derive an exact value for the asymptotic number of RNA shapes, by relying on an elegant relation between nonambiguous, contextfree grammars a ..."
Abstract

Cited by 13 (5 self)
 Add to MetaCart
RNA shapes, introduced by Giegerich et al. (17), provide a useful classification of the branching complexity for RNA secondary structures. In this paper, we derive an exact value for the asymptotic number of RNA shapes, by relying on an elegant relation between nonambiguous, contextfree grammars and generating functions. Our results provide a theoretical upper bound on the length of RNA sequences amenable to probabilistic shape analysis (37; 41), under the assumption that any base can basepair with any other base. Since the relation between contextfree grammars and asymptotic enumeration is simple yet not wellknown in bioinformatics, we give a selfcontained presentation with illustrative examples. Additionally, we prove a surprising 1to1 correspondence between πshapes and Motzkin numbers.
MultiChannel Image Identification and Restoration Using the ExpectationMaximization Algorithm
, 1995
"... Previous work has demonstrated the effectiveness of the ExpectationMaximization algorithm to restore noisy and blurred singlechannel images and simultaneously identify its blur. In addition, a general framework for processing multichannel images using singlechannel techniques has also been develop ..."
Abstract

Cited by 12 (5 self)
 Add to MetaCart
Previous work has demonstrated the effectiveness of the ExpectationMaximization algorithm to restore noisy and blurred singlechannel images and simultaneously identify its blur. In addition, a general framework for processing multichannel images using singlechannel techniques has also been developed. This paper combines and extends the two approaches to the simultaneous blur identification and restoration of multichannel images. Explicit equations for simultaneous identification and restoration of noisy and blurred multichannel images are developed, for the general case when crosschannel degradations are present. An important difference from the single channel problem is that the crosspower spectra are complex quantities, which further complicates the analysis of the algorithm. The proposed algorithm is very effective at restoring multichannel images, as is demonstrated experimentally. Subject terms: multichannel restoration, blur identification, EM algorithm 1 Introduction ...
Exact closedform performance analysis of optimum combining with multiple cochannel interferers and Rayleigh fading
 IEEE Trans. Commun
, 2003
"... Abstract—A new closedform expression is derived for the exact biterror probability (BEP) for optimum combining with binary phaseshift keying. The exact BEP expression is for multiple, equal power, cochannel interferers and multiple reception branches. It is assumed that the aggregate interference ..."
Abstract

Cited by 11 (2 self)
 Add to MetaCart
(Show Context)
Abstract—A new closedform expression is derived for the exact biterror probability (BEP) for optimum combining with binary phaseshift keying. The exact BEP expression is for multiple, equal power, cochannel interferers and multiple reception branches. It is assumed that the aggregate interference and noise is Gaussian and that both the desired signal and interference are subject to Rayleigh fading. The derivation starts by expressing the optimum combining decision statistic as a sum of quadratic forms of Gaussian random variables and it proceeds to average over the fading interference. The new BEP expression has low complexity as it contains only finite sums and products. Index Terms—Diversity reception, error probability performance, fading channels, interference suppression, optimum combining. I.