Abstract: A system (20) and method for managing change in software systems and applications manages change for both transient and persistent objects in a domain independent, non-intrusive, object-oriented fashion. The system (20) and method are designed and implemented as a combination of two abstract machines (22,24) consisting of a set of interface functions (25,26) visible to the application, and an internal or private state (27,28) which is hidden from such application. Applications use the interface functions (25,26) to obtain the services of this change management system (20) and method. Since the internal state of such system (20) and method is outside the application, change management is provided as a service without the need for altering or adding onto application data structures and interfaces.

Abstract: This is an automated speech recognition system, the system comprising: an input device for receiving voice signals; a means for computing the voice signals into stochastic RGDAGs and individual grammars; a search engine that directly processes the stochastic RGDAGs; a means for adding the individual grammars within the RGDAG; a means for replacing the individual grammars within the RGDAG; and a means for deleting the individual grammars within the RGDAG. The system can include a means for adding modified grammars within RGDAGs; a means for linking a terminal symbol from a first grammar to a start symbol in a second grammar; a means for finding a maximum depth of each symbol in the RGDAG; a means for determining if the system requires parse information for the grammar by looking for ancestor symbols; a means for specifing a plurality of RGs within the RGDAGs as starting points for the search engine; and a means for returning a parse through each of a plurality of RGs in the RGDAGs for a recognized utterance.

Abstract: In the present invention, an 8-bit input pixel gray level at a particular location in an image is screened, or compressed, down to k-bits where k is less than 8. A plurality of threshold tables are generated which represent a tone curve associated with the image in accordance with threshold table load logic (48). A preference matrix (42) which maps each pixel of the image to one of the threshold tables is generated using an image plane mapping interface (44) based on the input pixel location. An entry from the threshold table associated with the input pixel is selected using the gray level of the input pixel. The input pixel gray level is simultaneously compared with each of four thresholds in the selected entry using a quad comparator (58). A status register (60) is set based on the results for each comparison and is used to generate the k-bit output gray level associated with the input pixel.

Abstract: Simulation of a complex system begins by generating a plurality of system objects (12) describing items within the system through structural information within the system object (12). Each system object (12) corresponds to a plurality of parameter group objects (18, 20, 22), and a plurality of state objects (24, 26, 28) associated therewith. Each system object (12) also corresponds to a plurality of simulation objects (30, 32, 34) representing specific simulations of the complex system. Each simulation object (30) associates with a parameter group object (18) and a state object (24). Each object is indexed by an object identifier (OID) and parameter group and state objects are further indexed by their corresponding simulation object identifier. Structural, parameter, and state information are separated within their own individual objects allowing multiple threads of system configuration to be run through the simulation objects simultaneously for parallel comparison of simulation results.

Abstract: An improved system and method of determining the linearity of a ramp signal, including a ramp generator (11) for generating a ramp signal and a vernier delay (15) to provide scan stop signals where the samples are peak detected (17) and stored (21). The ramp generator (11) outputs are divided into equal sections controlled by a precise crystal controlled oscillator (25). The stored signals are processed and compared to detect errors.

Abstract: A method and system for implementing time scale modification wherein the method includes a Zero Crossing Module (22) for determining zero crossing points in the signal, a Feature Vector Module (24) for generating feature vectors describing the zero crossing points, a Distance Metric Module (26) for generating distance metrics describing local characteristics at the zero crossing points, an Alignment Module (28) for using the feature vectors and distance metrics for aligning and synchronizing the signal in accordance with local similarities and similarity over a selected time interval to generate a time scale modified signal. The present invention also includes a Cross Fade Module (20) for smoothing transitions between successive frames of the resulting time scale modified signal.

Abstract: The present invention is a speech processing system and method which includes a codebook generator (26) for generating a reference vector quantization codebook which describes a reference environment and for generating at least one secondary vector quantization codebook which describes at least one secondary environment. The secondary vector quantization codebooks are generated using the reference vector quantization codebook. A speech recognizer (38), which is trained using the reference vector quantization codebook, is also included. A pre-processor (32) accepts as input speech collected from an unknown environment and pre-processes that speech, before input to the speech recognizer (38), using the reference vector quantization codebook and an adaptation of the secondary vector quantization codebooks.

Abstract: This invention describes converting data representing wiring interconnect structures on an integrated circuit from traditional layout design databases which contain groups of unassociated geometries to object oriented databases which contain geometries with inherent connectivity information by reading all terminals (41), contacts (43), and unassociated geometries (45) and comparing (50, 54) end points of the geometric data with the pre-existing terminals and contacts initially read in. An internal terminal is created (52, 56) if no previously read-in terminal or contact exists.

Abstract: A machine and method for providing human voice sounding numbers includes storing in digital form segments of leading digit utterance, segments of trailing digit utterances, group pausing utterances and digit pair utterances. A data string of segments is read out of storage and concatenated.

Abstract: The memory and data management requirements for text independent speaker dependent recognition are drastically reduced by using a novel approach that eliminates the need for separate acoustic recognition models for each speaker. This is achieved by using speaker independent recognition models at the acoustic level. The speaker dependent data stored for each item to be recognized consists only of information needed to determine the speaker independent recognition model sequence for that item.

Abstract: A DMD display system includes an inverse gamma look-up-table (50) for converting raster scanned, gamma corrected video data of 8 bits to 12 bits inverse gamma data with 8 most significant bits (msb) and 4 least significant bits (lsb). The 8 msb are coupled to the micromirror of the DMD display (10) and the four lsb are delayed and halved such that one half of the lsb is added to the next pixel in the horizontal scan and one-half of the lsb is added to the next vertical pixel one line length delayed.

Abstract: A Karaoke system scoring method and system (10) is provided based on detecting, for example, frame energy (19 or 19') of the Karaoke singer and the frame energy of the original artist (29 or 29'). The frame energy is quantized (41 and 43) and compared (45) and based on the comparison a score (37) is generated and displayed (15).

Abstract: The present invention provides a system for a voice prompt talkover capability. The system and method include a speech recognizer (36), an echo canceler (24), a delay finder (27) and a utterance extraction mechanism (30). The delay finder (27) determines the propagation delay between outgoing signal and the incoming signal and stores the signal in a delay buffer (23) in accordance with the delay and the filter length of the echo canceler. The echo canceler (24), using the incoming signals, the delay buffer and the delay enhance the incoming signal by removing the echo. The utterance extractor (30) determines the start of an utterance based on energy in the vectors of the LPC buffer (28). Upon the detection of an utterance the utterance extractor (30) initiates the start of recognition and terminates operation of the echo canceler (24) and the prompt play out (14). A recognition result is returned from the speech recognition processor (36) to the application (12).

Abstract: A method involves capturing visual indicia as a digital image, isolating the visual indicia on the digital image, converting the visual indicia to image primitives, identifying and presenting the visual indicia. An apparatus contains a camera for capturing a visual image, a video-to-digital converter, and a computer with a processor, appropriate memory, and software.

Abstract: A method is provided for aligning a silicon wafer (20) in a fabrication tool (37) having a stage (22) involving the steps of producing a digital image of a portion of wafer (20) in a scope-of-view window (48), converting the digital image to image primitives, comparing the image primitives to grammar template primitives to locate a known intersection on wafer (20); and moving the stage (22) to align wafer (20). A method and apparatus are disclosed for determining the misregistration of two layers of a wafer (20) by converting targets (158, 160) to primitives and determining the relative displacement in symbolic space. The misregistration apparatus involves a camera (34), a video-to-digital converter (32), a computer (28), and a stage adjuster (24).

Abstract: A new signal quantization scheme is proposed which reduces fluctuation of the output signal by a signal quantizer (10) providing a quantized output signal and multiplying (18) said input signal X by a factor (1-w.sub.1) and finding a difference signal .DELTA..sub.in (11) between both the input signal and the previous input signal and multiplying (16) that by a weighting factor w.sub.2 from a control (13). The previous quantizer output signal Q.sub.prev is summed (17) with the weighted difference signal .DELTA..sub.in w.sub.2 and the sum is weighted by a weighting factor w.sub.1 at a multiplier (19) to yield w.sub.1 (Q.sub.prev +w.sub.2 .DELTA..sub.in). This signal is then summed at an adder (21) and applied to the quantizer (10) so that the quantizer is forced to match the fluctuation in the input signal as well as the signal itself.

Abstract: In a document creation stage (54), a document is created using a document generator (37) and then inspected, using a document analyzer (38), in an inspection stage (58). The document is created and inspected in accordance with defect types, defect severities and entry, reinspection and exit evaluation criteria data defined through a user interface (32) in an initialization stage (52). The defined data is stored to a definitions data storage (44) and, before the inspection stage (58) using a document analyzer (38), refined to the particular task document in a planning stage (56) using a refiner (36). During a logging stage (60), defect items are compiled and tallied to an inspection data storage (46) and used, in an inspection data analyzer (40), to make reinspection recommendations and, during a causal analysis stage (62), to generate recommended changes to the inspection processes. Documents are reworked in a rework stage (70) and final tallies are generated.

Abstract: An improved image scaling filter for a video display where a coefficient value for the closest input lines to a given output line that are less than two line lengths from the given output line are determined by cubic interpolation using the line distances. The input lines are multiplied by the coefficient for that line and the multiplied closest input line values are summed to determine the output line value.

Abstract: The method comprises the steps of converting the algorithm into a control flow graph having a plurality of nodes, each node representing at least one of the plurality of statements. The control flow graph further includes a conditional branching node for the data-dependent conditional branching statement. Reverse dominators are then computed for each of the plurality of nodes, from which a meet point node for each conditional branching node is derived. The method further provides for constructing a shadow symbol table housing the variables, and constructing a duplicate shadow symbol table for each possible value of the datum in response to the conditional branching node. The conditional branching node is also evaluated by assuming each possible value of the datum for each branch and assigning symbolic values to each of the variables in the shadow symbol table. The nodes contained in each branch are evaluated until the meet point node is reached. The duplicate shadow symbol tables are then merged.

Abstract: A method for symbolic evaluation of an algorithm is provided. The algorithm has at least one conditional branching statement dependent on the value of at least one input datum. The method comprises the steps of converting the algorithm into a control flow graph having a plurality of nodes, each node representing at least one of the plurality of statements. The control flow graph further includes a conditional branching node for the data-dependent conditional branching statement. Reverse dominators are then computed for each of the plurality of nodes, from which a meet point node for each conditional branching node is derived. The method further provides for constructing a shadow symbol table housing the variables, and constructing a duplicate shadow symbol table for each possible value of the datum in response the conditional branching node.