206 resultados para Poor free


Relevância:

20.00% 20.00%

Publicador:

Resumo:

Using an entropy argument, it is shown that stochastic context-free grammars (SCFG's) can model sources with hidden branching processes more efficiently than stochastic regular grammars (or equivalently HMM's). However, the automatic estimation of SCFG's using the Inside-Outside algorithm is limited in practice by its O(n3) complexity. In this paper, a novel pre-training algorithm is described which can give significant computational savings. Also, the need for controlling the way that non-terminals are allocated to hidden processes is discussed and a solution is presented in the form of a grammar minimization procedure. © 1990.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Advances in the dual electron-beam recrystallization technique arising from the fast scanning of a line beam parallel to the edges of narrow seeding windows are described. The resultant recrystallized layers are essentially defect-free, have good surface flatness, and cover large areas.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

This paper describes two applications in speech recognition of the use of stochastic context-free grammars (SCFGs) trained automatically via the Inside-Outside Algorithm. First, SCFGs are used to model VQ encoded speech for isolated word recognition and are compared directly to HMMs used for the same task. It is shown that SCFGs can model this low-level VQ data accurately and that a regular grammar based pre-training algorithm is effective both for reducing training time and obtaining robust solutions. Second, an SCFG is inferred from a transcription of the speech used to train a phoneme-based recognizer in an attempt to model phonotactic constraints. When used as a language model, this SCFG gives improved performance over a comparable regular grammar or bigram. © 1991.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

A description is given of experimental investigations in which free electromechanical oscillations are obtained for the first time in ferroelectric liquid crystals. © 1997 American Institute of Physics.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

It is shown experimentally that an elastic mechanical stress in a crystal structure is a necessary factor for the appearance of free oscillations of the director of a ferroelectric liquid crystal. Such a mechanical stress arises as a result of internal textural perturbations in the presence of regions with a different orientation of the director or is produced by external pressure applied to one of the cell plates in the appropriate direction. © 1999 American Institute of Physics.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

This paper presents a practical destruction-free parameter extraction methodology for a new physics-based circuit simulator buffer-layer Integrated Gate Commutated Thyristor (IGCT) model. Most key parameters needed for this model can be extracted by one simple clamped inductive-load switching experiment. To validate this extraction method, a clamped inductive load switching experiment was performed, and corresponding simulations were carried out by employing the IGCT model with parameters extracted through the presented methodology. Good agreement has been obtained between the experimental data and simulation results.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

A free-space, board-to-board, adaptive optical interconnect demonstrator has been developed. Binary phase gratings displayed on a Ferroelectric Liquid Crystal Spatial Light Modulator are used to maintain data transfer at 1.25Gbps, given varying optical misalignment © 2005 Optical Society of America.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

The scattering of sound from a point source by a Rankine vortex is investigated numerically by solving the Euler equations with the novel high-resolution CABARET method. For several Mach numbers of the vortex, the time-average amplitudes of the scattered field obtained from the numerical modeling are compared with the theoretical scaling laws' predictions. Copyright © 2009 by Sergey Karabasov.