Abstract: A method of making a speech recognition model database is disclosed. The database is formed based on a training string utterance signal and a plurality of sets of current speech recognition models. The sets of current speech recognition models may include acoustic models, language models, and other knowledge sources. In accordance with an illustrative embodiment of the invention, a set of confusable string models is generated, each confusable string model comprising speech recognition models from two or more sets of speech recognition models (such as acoustic and language models). A first scoring signal is generated based on the training string utterance signal and a string model for that utterance, wherein the string model for the utterance comprises speech recognition models from two or more sets of speech recognition models. One or more second scoring signals are also generated, wherein a second scoring signal is based on the training string utterance signal and a confusable string model.

Abstract: An apparatus that after a time period settable by the user automatically returns an off-hook telephone to an on-hook condition. This apparatus thus electrically hangs up a telephone handset that is inadvertently left off-hook and allows normal handling of outgoing and incoming calls at other telephone extensions and in some embodiments, even allows calls to ring-in to the telephone having the inadvertently off-hook handset. The apparatus also senses when the inadvertently off-hook handset is returned on-hook and rapidly restores the automatic-return-to-an-on-hook condition switch to normal operation.

Abstract: A configuration parameter programming system of a telephone is presented which is executable by an end user or service person. The configuration parameter programming system is accessed through the use of a dial memory register or other data transfer means, and a control means. In this way, the Dial Pulse Rate, Dial Pulse Make/Break Ratio, DTMF Duration, DTMF Inter Digit Delay, Pause Duration and other configuration parameters of a telephone are programmable after manufacture without requiring hardware modification.

Abstract: A method and apparatus for enabling one or more packets (e.g., CDMA packets, CDPD packets etc.) to be encapsulated into a cell (e.g., an ATM cell, a B-ISDN cell) for transmission and/or reception over a broadband network (e.g., an ATM Channel). An illustrative method includes building a basic control octet for a packet of N bytes, the basic control octet including (1) an additional control bit that indicates if a supplemental control octet accompanies the basic control octet, and (2) a length indicator based on N.

Abstract: A method and apparatus for the processing of magnetic recording data signals which provides for the cancellation of the effects of precursor intersymbol interference (ISI) in the data signals. According to one illustrative embodiment, an adaptive filter is used to produce a replica of the precursor ISI, which is then subtracted from a delayed version of the data signal. The amount of the delay is advantageously equal to the delay between the main peak of the channel impulse response and the leading undershoot (i.e., the precursor ISI effect) which is to be canceled. The signal which is provided to the adaptive filter (to produce the precursor ISI replica) may be based on tentative decisions generated by a symbol-by-symbol detector, and may first be converted from three levels to two levels by a three-to-two level converter.

Abstract: High speed testing of a digital circuit may be performed although the rated frequency of the circuit exceeds the frequency capability of the test equipment. A digital circuit may be designed such that a controllable delay may be introduced in the timing paths of the circuit during testing using test stimuli which are applied at a clock rate that is less than the rated frequency of the circuit. By adding delay to the combinational signal path, testing of the circuit for operation at the maximum operating frequency is achieved during testing at a clock rate which is within the capability of the test equipment. The controllable delay may be incorporated as a delay element into a single-clock circuit and controlled by manipulation of the duty-cycle of a clock waveform which is applied to the circuit. The delay circuit is so designed that its function is also testable. In a multi-clock circuit, the delay is added to the circuit by skewing one clock signal with respect to the other clock signals.

Abstract: The present invention discloses an improvement to a wireless packet data communication system by permitting a subscriber using a mobile end system (M-ES) to select the inter-exchange carrier (IXC) which will carry the data over those network paths for which the subscriber bears the cost during a particular communication session. The present invention allows the subscriber to specify a preferred inter-exchange carrier in an optional preferred IXC field added to an end system hello message. During registration and authorization procedures, an identifier corresponding to the preferred inter-exchange carrier is transmitted between a serving mobile data intermediate system (MD-IS) and a home MD-IS in a preferred IXC field added to standard registration protocol messages.

Abstract: A method and apparatus for control of lasing wavelength in distributed feedback lasers using an improved waveguide. The waveguide comprises a periodic structure or grating and an active stripe which is oriented at a nonzero angle .THETA. with respect to a line perpendicular to the lines of the grating. Incremental changes in the operating wavelength of a plurality of lasers can be achieved by forming the active stripe of each waveguide of each laser at a different angle .THETA.. This method and apparatus provides a fine degree of control over the operating wavelength of distributed feedback lasers and is particularly useful in high capacity optical communications networks.

Abstract: A novel method and apparatus is disclosed for controlling inverse discrete cosine transform ("IDCT") mismatch between an encoder and decoder using different IDCT implementations under all coding conditions. Sychronization between the encoder and decoder is forced by selectively setting certain small DCT coefficients, which are prone to mismatch, to zero at the encoder. Advantageously, the invention may be implemented solely in the encoder to minimize system complexity and cost, and allow for IDCT mismatch control utilizing conventional decoders.

Abstract: A memory sense amplifier for a static random access memory includes a pair of transistor amplifiers respective to the bit lines threading the memory. The power consumed is minimized without sacrificing speed of operation by temporarily connecting the source electrodes of the transistor amplifiers to the bit lines to allow them to track the states of the bit lines before a current path is completed to the drains of the transistors to allow them to draw current from the bit lines, thereby minimizing the time that the sense amplifiers are permitted to draw current from the bit lines. In addition, an economy of circuitry is achieved by eliminating the need for a separate latch circuit by disconnecting the sense amplifiers from the bit lines and thereafter enabling them to latch the information state read from the bit lines.

Abstract: A telephone terminal that allows simple word commands to produce different results depending on the identity of the caller. The use of a limited set of word commands allows the utilization of an inexpensive voice recognition subsystem. A first single word command results in a caller from a first group of people being immediately placed on a speakerphone; whereas, a caller from a second group of people receives the audio message "Please stay on the line a few seconds, and I will personally answer your call". Further, for the same audio command, the called party can pre-record an audio message for the first group; whereas the second group always receives a voice synthesized message. A second single word command results in a caller from the first group of people receiving an audio message that states "The name of the caller, I will call you in five minutes". Whereas, a caller from the second group of people receives the audio message "Please call back in five minutes".

Abstract: The present invention is an apparatus and method for increasing the density of run-length-limited (RLL) block codes without increasing error propagation. By inserting a number of uncoded bytes (M) between each coded byte, the coding density is thereby increased. Starting with an RLL code with a block length (I) which is, for example, a multiple of 8, a number (M) of uncoded bytes may be inserted between each coded byte. The resulting density is: (I+8M)/(J+8M), wherein the resulting k constraint, of the (d,k,l) constraints is increased by 8M, and the resulting l constraint is increased by 4M. For example, starting with an RLL code having a coding density of 8/9 (I=8, J=9) and constraint set of (0,4,4), inserting one uncoded byte between each coded byte (M=1) results in a coding density of 16/17 which is 5.88% greater than the original 8/9 coding density. The constraint set is also increased to (0,12,8), where k is increased by 8 and l is increased by 4.

Abstract: Optical fiber support members (30 and 31) are formed by making an array of V-grooves (25) in a monocrystalline material (26), depositing metal (27) on the array, and removing the metal member to provide an insert. The insert is used to form the support members by injection molding.

Abstract: Door-to-door transportation is provided in a nearly on-demand manner by employing a set of multipassenger vehicles, e.g., vans, whose routes are continuously updated. The position of each multipassenger vehicle, as well as the traffic conditions along the projected route of each multipassenger vehicle, are constantly monitored. Upon receiving a request for a ride, e.g., via telephone, which specifies a) the pick-up location, b) the destination and, optionally, c) a desired time of arrival thereat and d) the number of people, a suitable "path" to the destination which satisfies the request is determined, if possible. A "path" for a ride is a specification of the particular multipassenger vehicles, and the location and nature of any required transfers for a person's requested ride. A path may also include the streets over which the rider will pass and any scheduled stops.

Abstract: In an improvement over conventional finite element techniques, a prolate spheroidal infinite element is used for the modeling of acoustic fields in exterior, fluid-filled domains surrounding a structure. This prolate infinite element is based on a multipole expansion that describes, to arbitrary accuracy, any scattered and/or radiated field exterior to a prolate spheroid of any eccentricity. The prolate infinite elment is readily incorporated in any structural or acoustical finite element code. Structural acoustic modeling with the element is several orders of magnitude faster than modeling, at comparable accuracy, with the well-known and widely used boundary element method, at least for large-scale problems involving tens of thousands of degrees of freedom.

Abstract: In an improvement over conventional finite element techniques, an oblate spheroidal infinite element is used for the modeling of acoustic fields in exterior, fluid-filled domains surrounding a structure. This oblate infinite element is based on a multipole expansion that describes, to arbitrary accuracy, any scattered and/or radiated field exterior to an oblate spheroid of any eccentricity. The oblate infinite elment is readily incorporated in any structural or acoustical finite element code.

Abstract: A hybrid type of decision feedback equalizer (DFE) structure that approaches optimal DFE performance is described. The hybrid DFE includes both an intersymbol interference DFE (ISI-DFE) and a noise predictive DFE (NP-DFE). In particular, an adaptive feedforward filter section is followed by both an intersymbol interference (ISI) predictive filter and a noise predictive (NP) filter. The corresponding transmitter uses a precoder that utilizes optimal coefficient values, denoted as C(z), of the hybrid DFE structure. In particular, C(z) is equal to (1+I(z)) (1+N(z))-1, where, the notation I(z) and N(z), as known in the art, represent the resulting coefficient values after adaptation of the ISI-DFE and the NP-DFE, respectively.

Abstract: An optical integrated circuit provides multiple power splitting functions and comprises a plurality of optical power splitters networked together. In one embodiment, the optical integrated circuit has two 2.times.2 optical power splitters ganged together such that the circuit simultaneously provides two different 1.times.2 power splitting functions and four different 1.times.3 power splitting functions. In another embodiment, the optical integrated circuit has three 2.times.2 optical power splitters ganged together such that the circuit simultaneously provides two 1.times.2 power splitting functions, four 1.times.3 power splitting functions, and two 1.times.4 power splitting functions. Since each integrated circuit provides more than one power splitting function, the integrated circuit is more versatile than a single power splitter.

Abstract: There is disclosed a digital-to-analog converter including a resistor string having intermediate taps at resistor junctions as well as resistor-potential junctions. Switching transistors are coupled between a respective intermediate tap and an output node. Decode circuits are capable of switching at least two transistors to be in the on state at the same time to electrically couple more than one intermediate tap to the output node to produce at least one analog output. In one embodiment, one row select line can be energized simultaneously with at least two column select lines. Alternatively, at least two row select lines can be energized simultaneously with one column select line. A DAC in accordance with the present invention is suitable for fabrication in the form of a monolithic integrated circuit and requires less area while maintaining the same degree of resolution as prior art DACs. The DAC also may be used with successive approximation circuitry to provide an analog-to-digital converter.