Abstract: A technique to improve the recognition accuracy when transcribing speech data that contains data from a wide range of environments. Input data in many situations contains data from a variety of sources in different environments. Such classes include: clean speech, speech corrupted by noise (e.g., music), non-speech (e.g., pure music with no speech), telephone speech, and the identity of a speaker. A technique is described whereby the different classes of data are first automatically identified, and then each class is transcribed by a system that is made specifically for it. The invention also describes a segmentation algorithm that is based on making up an acoustic model that characterizes the data in each class, and then using a dynamic programming algorithm (the viterbi algorithm) to automatically identify segments that belong to each class. The acoustic models are made in a certain feature space, and the invention also describes different feature spaces for use with different classes.

Abstract: Silicon nitride is etched employing a composition containing a fluoride containing compound, certain organic solvents, and water.

Abstract: A computer system includes a node interconnect to which at least a first processing node and a second processing node are coupled. The first and the second processing nodes each include a local interconnect, a processor coupled to the local interconnect, a system memory coupled to the local interconnect, and a node controller interposed between the local interconnect and the node interconnect. In order to reduce communication latency, the node controller of the first processing node speculatively transmits request transactions received from the local interconnect of the first processing node to the second processing node via the node interconnect, where each such request transaction specifies an associated datum. The node controller of the second processing node handles each speculatively transmitted request transaction received in response to a directory state of its associated datum.

Abstract: A non-uniform memory access (NUMA) computer system includes an interconnect to which multiple processing nodes (including first, second, and third processing nodes) are coupled. Each of the first, second, and third processing nodes includes at least one processor and a local system memory. The NUMA computer system further includes a transaction buffer, coupled to the interconnect, that stores communication transactions transmitted on the interconnect that are both initiated by and targeted at a processing node other than the third processing node. In response to a determination that a particular communication transaction originally targeting another processing node should be processed by the third processing node, buffer control logic coupled to the transaction buffer causes the particular communication transaction to be retrieved from the transaction buffer and processed by the third processing node.

Abstract: A liquid crystal mode includes a layer of a liquid crystal material disposed between homeotropic alignment layers having a parallel alignment direction, the liquid crystal material including a negative dielectric anisotropic material. The liquid crystal material having a first state characterized by a vertical alignment of molecules which is substantially perpendicular to the alignment layers, and a second state which is characterized by the orientation of the molecules of the liquid crystal material which is oriented responsive to an electric field such that the molecules above a mid-plane of the liquid crystal material have a tilt in a first direction and the molecules below the mid-plane have a tilt opposite the first direction while the molecules adjacent the alignment layers and at the mid-plane remain substantially perpendicular to the alignment layers. The mid-plane is substantially parallel to the alignment layers.

Abstract: A collar oxide is formed in a provided a semiconductor substrate having (3) a partially full trench, (2) (i) fill surface defined by fill material partially filling said trench, (ii) upper surface outside of said trench, and (iii) trench sidewall surface not covered by said fill material, and (3) a conformal oxide layer overlying said fill, upper, and sidewall surfaces, by selectively etching as follows:(a) contacting the substrate with a mixture of hydrogen-containing fluorocarbon and an oxygen source under reactive ion etching conditions until at least a portion of the conformal oxide layer on the upper surface is removed, and(b) contacting the substrate from step (a) with a mixture of a hydrogen-free fluorocarbon and a diluent gas under reactive ion etching conditions to further remove conformal oxide remaining on the fill surface and to overetch the upper and fill surfaces, whereby a substantial portion of conformal oxide remains on the side walls to form the collar oxide.

Abstract: Object-code instruction traces are employed to analyze selected instructions of an application program for possible failure when confronted by a year-2000 date. The analysis includes directly identifying one or more instructions of the application program that may fail, as well as identifying whether the one or more instructions have a characteristic of a predefined false-positive failure pattern. A failure-pattern descriptor is assigned to each examined instruction which is indicative of whether the instruction may fail when confronted by a date in the year-2000 range, and whether the instruction is a possible false-positive failing instruction. The analysis employs user-specifiable run-control values, as well as predetermined filter-specification values in comparing traces of each selected object-code instruction to predefined instruction failure patterns.

Abstract: A method for increasing chip yield by reducing black silicon deposition in accordance with the present invention includes the steps of providing a silicon wafer suitable for fabricating semiconductor chips, depositing a first layer over an entire surface of the wafer, removing a portion of the first layer to expose a region suitable for forming semiconductor devices and etching the wafer such that a remaining portion of the first layer prevents redeposition of etched material on the wafer. A semiconductor assembly for reducing black silicon deposition thereon, includes a silicon wafer suitable for fabricating semiconductor chips, the wafer having a front surface for forming semiconductor devices, a back surface and edges. A deposited layer is formed on the wafer for covering the back surface and the edges such that redeposition of silicon on the back surface and edges of the wafer during etching is prevented.

Abstract: The present invention prevents a head/slider assembly fixed on a flexure of a head suspension assembly from having an undesired pitching motion when an unexpected excess shock is applied to a hard disk device when the head suspension is at the best position of the unload ramp.

Abstract: A computer system having a pipelined computer processor, which executes a relatively simple instruction set in a hardware controlled execution unit and executes a relatively complex instruction set in a milli-mode architected state with a millicode sequence of simple instructions in said hardware controlled execution unit, a millicode operating in a milli-mode state when macro-mode decoding by said processor is suspended to cause the system to subsequently use processor milli-registers and the processor's decoder decodes them and schedules them for execution upon entry into the processor milli-mode. Millicode flags allow specialized update and branch instructions and flags are either cleared or specifically set for a millicode instruction. A millicode instruction for editing functions processes one byte of an input pattern string, generates one byte of an output string, and updates various pointers and state indications to prepare for processing the next byte in a string.

Abstract: A method and apparatus for controlling a spindle motor having at least one winding in a storage device are proposed, the storage device having at least one data storage medium rotatably supported by the spindle motor and having a plurality of tracks on each surface thereof including data regions and nondata regions, and a transducer assembly for selectively positioning a transducer over each data storage surface of the storage medium for reading and writing of information stored on the medium. The method and apparatus control rotation of the spindle motor by providing a voltage waveform to each winding at selected switching times during each rotation such that switching does not occur while the head is positioned over data. In a first embodiment, switching during nondata regions is assured by setting the switch points to coincide in time with a fixed data storage surface format. Alternatively, the data storage surface is written with servo sectors to coincide in time with preselected switch points.

Abstract: A method, memory, and apparatus, having at least a processor, memory, and touchscreen, for creating a virtual pointing device. The method includes the steps of displaying a non-activated virtual pointing device on the touchscreen, in response to detecting at least one finger placed on the virtual pointing device, activating the virtual pointing device, and moving the virtual pointing device in accordance with movement of the at least one finger.

Abstract: An apparatus and method for correcting asymmetry in the duty cycle of a differential signal is disclosed. The differential receiver circuit includes a first converter including a positive input coupled to a first direct current (DC) blocking capacitor, a negative input coupled to a second DC blocking capacitor, and an output resistively coupled to the positive input of the first converter. The differential receiver circuit further includes a second converter including a positive input coupled to the negative input of the first converter, a negative input coupled to the positive input of the first converter, and an output resistively coupled to the negative input of the first converter. The respective outputs of the first and second converters are preferably inverting outputs.

Abstract: An apparatus and method, for use with a data processing system having an architecture, which provide for the possibility of smart card use without unduly impacting the data processing system's architecture. An enclosure, having at least one PC card acceptor and at least one smart card acceptor, is created. A suspension mechanism is operably connected to enclosure in a fashion such that suspension mechanism is capable of physically connecting enclosure to a motherboard such that enclosure is suspended above a component of motherboard. The method and system give rise to several advantages. One advantage is that board space is saved in that the enclosure is suspended over existing board components. A second advantage is that the method and system allow original equipment manufactures to make the provision of smart card usage optional, in that smart card usage can be provided by a simple add-on to lower-end systems.

Abstract: Various services within a computer system are automatically set up as users are added to the system through a directory service. A user template is selected that already contains default settings for each of the services available on a system such as e-mail and a web service. Various user templates exist and contain different default settings for each available service depending upon the user. Default settings from each service are copied onto a new user form. Next, default settings from the selected user template are copied to the new user form and override any conflicting default settings from the services. The default settings from the new user form are used to automatically set up and enable each service for the new user. The data from the new user form is copied into a central account database that contains rows of user information, with columns corresponding to a particular service. Each user is automatically enabled for use with available services including an e-mail account and a generated web page.

Abstract: A system and method for controlling client access to enterprise resources through a middle tier server. Enterprise resource authorizations are maintained in a middle tier server. Users authenticate with the server causing it to map and transform the client access authorization into enterprise resource credentials. Enterprise resources are accessed after authorizing using the transformed credentials.

Abstract: A method and apparatus for generating a Common Gateway Interface (CGI) adaptor customized for a stored procedure. In accordance with the present invention, stored procedure application developers can provides their stored procedure to a builder, which creates a CGI adaptor customized for the stored procedure. To generate the CGI adaptor, the builder parses the stored procedure to identify the arguments of the stored procedure. Then, using the stored procedure arguments, the builder generates logic to extract variables from an input data stream, wherein the variables correspond to the arguments of the stored procedure. The builder also generates logic to invoke the stored procedure using the variables extracted from the input data stream. The generated logic comprises the customized CGI adaptor.

Abstract: An apparatus for retrieving information in a computer network system having at least one network server device and a user device. The user device and said network server having established a successful network session so that they are in processing communication with one another. At first a target page is displayed on a screen or monitor connected to the user device. The target page including a number of first level link references for selection by any user where each first level link reference further includes subsequent level link references. An on-screen search option menu will then be provided to allow any user to search all reference links provided by the target page and all subsequent pages, and to identify only specific references according to the menu selection made. Some of the menu selection provided according to different embodiments of the present invention are media searches, FTP file searches, HREF searches and usernet/newsgroup searches.

Abstract: According to the preferred embodiment, a method is provided for automatically coloring VLSI design elements for the purpose of assigning binary properties to the elements. The preferred method is particularly applicable for use generating phase shift mask designs from VLSI CAD datasets. The preferred method uses net coloring to automatically generate a data set of colored elements. The preferred method is not dependent on the order in which the elements are operated upon. The preferred method has the additional advantage of being able to automatically detect conflicts that prevent the VLSI design from being optimally colored. The preferred method is equally applicable to hierarchical VLSI databases with nested components and traditional flat databases. When applied the hierarchical databases, the preferred method provides element coloring with minimal data flattening required.

Abstract: Method of improving the resistance of a metal against degradation from exposure to fluorine released from a fluorine-containing material by forming a fluorine-free silicon dioxide barrier layer between the insulator material and the metal. The invention is especially useful in improving corrosion and poisoning resistance of metallurgy, such as aluminum metallurgy, in semiconductor structures. The invention also covers integrated circuit structures made by this method.