Abstract: A document marker, including first values dependent upon the layout and the contents of the document and assigned by generating or preprocessing software, is provided in machine-readable symbology on the face of a printed version of the document. The marker may include encoded document layout information and values assigned on sequences of the original text, including text-dependent decimation sequences, error correction codes or check-sums. Upon optical character recognition scanning, or other digitizing reproduction, the marker is also scanned. The scanning computer, having corresponding software, assigns second values dependent upon the layout and contents of the reproduced document. Upon comparison of the first and second decimation sequences, line and character errors can be detected and some errors corrected, thereby generating re-aligned candidate sequences.

Abstract: A system and method for storage and retrieval of pictographic images as computer file names. Upon generation of an input image, the computer saves the image and extracted spatial and temporal information about the generation of the image. The extracted information can be accessed for comparison, ordering and retrieving of stored pictographic file names and subsequent display to the user. Final selection of the desired name is performed visually by the user while viewing a display of the ordered file names via a document browser.

Abstract: Apparatus for determining a distance between two handwritten strings in a database. A processor extracts global features from each string. The processor divides the string into strokes, and identifies a plurality of bounding boxes. Each box contains a different stroke. The processor extracts global features from the suing, including: (1) a number of points; (2) a maximum angle between a first point in the string and a corner of the tallest bounding box; (3) a number of positive inversions; and (4) a number of negative inversions. The apparatus calculates the distance between the strings based on all of the numbers of points, maximum angles, numbers of positive inversions and numbers of negative inversions. A fixed query tree index may be formed. The tree has leaves and internal nodes belonging to multiple levels. A different key is associated with each level. Each key is a handwritten string.

Abstract: A method and system are provided for maintaining coherency between a server processor and a client processor that has a cache memory. The server may, for example, be a fixed location mobile unit support station. The client may, for example, be a palmtop computer. The server stores a plurality of data values, and the client stores a subset of the plurality of data values in the cache. The server processor periodically broadcasts invalidation reports to the client processor. Each respective invalidation report includes information identifying which, if any, of the plurality of data values have been updated within a predetermined period of time before the server processor broadcasts the respective invalidation report. The client processor determines, based on the invalidation reports, whether a selected data value in the cache memory of the client processor has been updated in the server processor since the selected data value was stored in the cache memory.

Abstract: A certificate, calculated on the text of a document to be reproduced, is included in the document in a form which is machine-readable. Upon optical character recognition scanning, or other digitizing reproduction, the certificate is also scanned and stored for comparison. A new certificate is calculated on the contents of the reproduced document and compared to the scanned certificate. If the two certificates are different, it is known that at least one error exists in the reproduced document. More than one certificate can be provided for a document in order to facilitate error location on the document and error correction therein.

Abstract: A speaker normalization method is described based on spectral shifts in the auditory filter domain. The method is characterized by using an estimated vocal tract length as a criterion to determine the spectral shift value. Certain constraints are found to be necessary for the shift in the auditory filter domain, and two techniques based on these constraints, the One-Bark shift and the refined Bark-scale shift, are introduced. When tested in vowel classification experiments, significant performance improvement was obtained for both techniques. The method is useful for speaker normalization in speaker-independent speech recognition.

Abstract: Apparatus and method for metal line deposition on a substrate. Laser ablation of a metal film coated on a first substrate removes metal ions from the film. The ions travel forward to a surface of a second substrate disposed opposite the metal film on the first substrate and are deposited on the second substrate. A positive electrode on the first substrate, a negative electrode on the second substrate, and a power supply create an electric field that is simultaneously applied across the first and second substrates. The positively charged ions in the second substrate migrate toward the negative electrode of the electric field, leaving a surplus of negatively charged ions at the surface of the second substrate. The negative ions in the second substrate electrostatically bond with the positively charged ablated metal ions, thereby assisting the bonding of metal lines to the substrate.

Abstract: A multipass recognition strategy selects the N-best hypotheses resulting from each pass and propagates these N-best to the next pass. This strategy outperforms conventional hidden Markov model recognizers using a grammar constraining all possible names. Real time recognition of continuously spelled names is made feasible, in part, because the processor-intensive costly constraints are applied, if at all, in the 4th pass, after the system has produced a much smaller dynamic grammar.

Abstract: A self-learning speaker adaptation method for automatic speech recognition is provided. The method includes building a plurality of Gaussian mixture density phone models for use in recognizing speech. The Gaussian mixture density phone models are used to recognize a first utterance of speech from a given speaker. After the first utterance of speech has been recognized, the recognized first utterance of speech is used to adapt the Gaussian mixture density hone models for use in recognizing a subsequent utterance of speech from that same speaker, whereby the Gaussian mixture density phone models are automatically adapted to that speaker in self-learning fashion to thereby produce a plurality of adapted Gaussian mixture density phone models.

Abstract: A system and method for reducing signal-to-noise ratio in a multi-component analog signal during analog to digital conversion. The multi-component analog signal is made up of a plurality of separate analog signals, with each of the separate analog signals having separate amplitude levels. A reference gain control signal, corresponding to the largest amplitude component of the multi-component signal, is first determined. Next, the reference gain control signal is used to generate a gain control signal. The gain control signal is then used to adjust each component of the multi-component signal by a gain value. Finally, the gain adjusted signals are converted to corresponding digital signals.

Abstract: A method for indexing a plurality of handwritten objects is provided. A B-tree data structure of order m is generated, where m is an integer. The B-tree has a plurality of nodes divided into a plurality of levels ordinally numbered 0.sup.th through n.sup.th. Each node in the 0.sup.th level is a leaf. Each node in the 1.sup.th level has at least m/2 leaves as children. Each one of the handwritten objects is assigned to a respective leaf. A respectively different hidden Markov model (HMM) is associated with each respective child of each of the nodes in the 1.sup.th to n.sup.th levels. Each one of the nodes in the 1.sup.th to n.sup.th levels contains the respective HMM associated with the child of the one node. Each HMM in each one of the nodes in the 1.sup.th level is trained to accept the handwritten object of the respective leaf that is a child of the one node. Each HMM associated with any of the nodes in the 2.sup.th through n.sup.

Abstract: The dictionary is broken into clusters by first grouping the dictionary according to a rule based procedure whereby the dictionary is sorted by word length and alphabetically. After sorting, a plurality of first cluster centers is generated by selecting the dictionary entries that differ from neighboring entries by the first letter. Each of the dictionary entries is then assigned to the closest one of the first cluster centers using a dynamic time warping procedure. These newly formed clusters are then each analyzed to find the true cluster center and the dictionary entries are then each assigned to the closest true cluster center. The clusters, so formed, may then be rapidly searched to locate any dictionary entry. The search is quite efficient because only the closest cluster to the desired dictionary entry needs to be searched.

Abstract: The device detects the beginning and ending portions of speech contained within an input signal based on the variance of frequency band limited energy within the signal. The use of the variance allows detection which is relatively independent of an absolute signal-to-noise ratio with the signal, and allows accurate detection within a wide variety of backgrounds such as music, motor noise, and background noise, such as other speakers. The device can be easily implemented using off-the-shelf hardware along with a high-speed special purpose digital signal processor integrated circuit.

Abstract: A filter selector generates a filter address value to select one of a number of frequency characteristics of a filter to resample a first data signal from a first segment of data in order to generate second sampled data signal in a second segment of data. The filter selector includes an adder which adds a resampling factor to a previous address value to generate a current address value if the previous address value is less than a comparator value and which subtracts the comparator value from the previous address value to generate the current address value if the previous address value is greater than or equal to the comparator value. The current address value is shifted by n bits, where n is an integer, to generate the filter address value.

Abstract: An analog to digital converter which includes a resistive network for generating a number of first reference voltages related to each other by a first linear relationship and a number of second reference voltages related to each other by a second linear relationship, where the first linear relationship is different from the second linear relationship. Also included are first comparators which compare an analog signal to each one of the first reference voltages to produce specified first comparator signals where a respective comparator of the first comparators is provided for each of the first reference voltages. Second comparators are provided to compare the analog signal to each of the second reference voltages to produce second comparator signals where a respective comparator of the second comparators is provided for each one of the second reference voltages.

Abstract: A system and method for the transmission and recovery of asynchronous data. For instantaneous synchronization, the receiver is equipped with a high frequency timing base which has a far higher frequency than either the data-generating or transmitting rate. The receiver clock is instantly synchronized upon detection of the first transition of the incoming data packet. Data packet verification can be conducted and data processed with minimal loss of data.

Abstract: An apparatus for controlling equipment from a remote location is disclosed. A control keypad having a distinctive key arrangement such as an array is used to interact with a display device (e.g. TV), located independently from the control keypad. A control menu is displayed on the display device where the control menu has a background image which substantially matches the distinctive arrangement of the keypad, and a foreground image which is overlaid on the background image to associate menu options with each actuator on the keypad. Finally, responsive to the option selected by the keypad, the corresponding function is executed. The menu system is implemented in the context of a home-bus system such as CEBus.

Abstract: The invention relates to general purpose interprocessor communication implemented through a distributed shared memory network connecting a plurality of processors, computers, multiprocessors, and electronic and optical devices. The invention teaches an apparatus for shared memory based data transfer between a multiplicity of asynchronously operating devices (processors, computers, multiprocessors, etc.) each using possibly distinct memory address translation architectures. The invention further teaches shared virtual memory network communication and administration based on a unique network memory address translation architecture. This architecture is compatible with and augments the address translation and cache block replacement mechanisms of existing devices.