Abstract: A method and system for adaptively filtering a speech signal in order to suppress noise in the signal. The method includes decomposing the signal into multiple frequency subbands, each having a center frequency, estimating a signal-to-noise ratio for each subband, and providing multiple filters, each filter designed for one of a number of selected signal-to-noise ratio independent of the center frequencies of the subbands. The method also includes selecting a filter for filtering each subband, where the filter selected depends on the signal-to-noise ratio estimated for the subband, filtering each subband according to the filter selected, and combining the filtered subbands to provide an estimated filtered speech signal. The system includes appropriate hardware and software for performing the method.

Abstract: A multi-actuator system for active sound and vibration cancellation utilizes an LMS type algorithm having an adaptive filter. However, the error signal rather than the input is filtered through an adjoint filter of the error channel to drive an adaptive filter which in turn drives, for example, a loudspeaker to provide destructive interference for noise cancellation. The adjoint filter is realized by converting a standard filter's flow direction, such as a finite impulse response filter, swapping branching points with summing junctions and unit delays with unit advances. For multiple-input-output systems, computational complexity is significantly reduced.

Abstract: A method and system for generating an estimate of a clean speech signal extracts time trajections of short-term parameters from a noisy speech signal to obtain a plurality of frequency components each having a magnitude spectrum and a phase spectrum. The magnitude spectrum is then compressed, filtered and then decompressed to obtain a modified magnitude spectrum. The speech signal is then reconstructed using the original phase spectrum and the modified magnitude spectrum.

Abstract: A binary, boron-based alloy as a source for field-emission-type, ion-beam generating devices, wherein boron predominates in the alloy, preferably with a presence of about 60 atomic percent. The other constituent in the alloy is selected from the group of elements consisting of nickel, palladium and platinum. Predominance of boron in these alloys, during operation, promotes combining of boron with trace impurities of carbon in the alloys to form B.sub.4 C and thus to promote wetting of an associated carbon support substrate.

Abstract: A substrate for the growth of monocrystalline .beta.-SiC is formed by providing a body of monocrystalline hexagonal material having a planar surface with a lattice parameter that is within .+-.5% of the lattice parameter of 6H.alpha.-SiC in the basal plane and growing a body of monocrystalline cubic material on the surface to provide a planar cubic material surface that is without grain boundaries, subgrain boundaries, double positioning boundaries, and pits. The cubic material, for example TiC, ZrC, HfC, or TiN, has a rock salt structure and a lattice parameter within .+-.5% of the lattice parameter of .beta.-SiC. Monocrystalline .beta.-SiC can be nucleated and grown on the surface of the cubic material.

Abstract: Improved system and method for speaker-independent speech token recognition are described. The system is neural network-based and involves processing a sequence of spoken utterances, e.g. separately articulated letters of a name, to identify the same based upon a highest probability match of each utterance with learned speech tokens, e.g. the letters of the English language alphabet, and based upon a highest probability match of the uttered sequence with a defined utterance library, e.g. a list of names. First, the spoken utterance is digitized or captured and processed into a spectral representation. Second, discrete time frames of the DFT are classified phonetically. Third, the time-frame outputs are used by a modified Viterbi search to locate segment boundaries, near which such segment boundaries lies the information that is needed to discriminate letters. Fourth, the segmented or bounded representation is reclassified using such information into individual hypothesized letters.

Abstract: A bright, short wavelength blue-violet phosphor for electroluminescent displays comprises an alkaline-based halide as a host material and a rare earth as a dopant. The host alkaline chloride can be chosen from the group II alkaline elements, particularly strontium chloride (SrCl.sub.2) or calcium chloride (CaCl.sub.2), which, with a europium (Eu) or cerium (Ce) rare earth dopant, electroluminesces at a peak wavelength of 404 and 367 nanometers (nm) respectively. The resulting emissions have CIE chromaticity coordinates which lie at the boundary of the visible range for the human eye thereby allowing a greater range of colors for full color flat panel electroluminescent (FPEL) displays.

Abstract: A substrate for the growth of monocrystalline .beta.-SiC is formed by providing a body of monocrystalline hexagonal material having a planar surface with a lattice parameter that is within .+-.5% of the lattice parameter of 6H.alpha.-SiC in the basal plane and growing a body of monocrystalline cubic material on the surface to provide a planar cubic material surface that is without grain boundaries, subgrain boundaries, double positioning boundaries, and pits. The cubic material, for example TiC, ZrC, HfC, or TiN, has a rock salt structure and a lattice parameter within .+-.5% of the lattice parameter of .beta.-SiC. Monocrystalline .beta.-SiC can be nucleated and grown on the surface of the cubic material.

Abstract: The apparatus of the present invention includes a light source for producing a coherent beam of light. A beam splitter is provided for splitting the beam of light into a first, transmitted beam segment and a second, local oscillator beam. A mechanism is provided for frequency shifting the frequency of one or both beam segments and for directing the first beam segment to a target. A remote target for scattering the first beam is provided. A mechanism is provided for combining the scattered first beam segment returning from the target and the second beam segment into a combined beam, and detecting the combined beam. A detector is operative to generate a signal indicative of the crosswind along the path of the directed first beam segment. A mechanism for determining the wind speed normal to the path from the generated signal is also provided.

Abstract: A method for modulating a liquid metal ion source includes generating an ion beam, directing a light beam at the liquid metal ion source and inducing a modulation in the ion beam by modulating the light beam.An apparatus for carrying out the invention includes a light beam source for generating a light beam, a modulator for modulating the light beam and a mechanism for directing the modulated light beam at the light metal ion source.