Abstract: Methods, systems, and programming for producing and displaying a scaled-down representation of subpixel-optimized images and digital content including such images that is represented by a mark-up language. Some embodiments display the text in such digital content using a different subpixel optimization than that used for the images. Others produce the subpixel-optimized images by calculating the luminosity of a subpixel as a function of the length of a plurality of coverage lines within a window in a source image corresponding to the subpixel that is covered by source image pixels having the subpixel's color. Some embodiments calculate the luminosity of a subpixel in such a subpixel-optimized image as a function both of the average luminosity of pixels in the subpixel's source image window and as a function of any color balancing distribution between resulting subpixel luminosities necessary to reduce color imbalance.

Abstract: A handheld device with both large-vocabulary speech recognition and audio recoding allows users to switch between at least two of the following three modes: (1) recording audio without corresponding speech recognition; (2) recording with speech recognition; and (3) speech recognition without audio recording. A handheld device with both large-vocabulary speech recognition and audio recoding enables a user to select a portion of previously recorded sound and have speech recognition performed upon it. A system enables a user to search for a text label associated with portions of unrecognized recorded sound by uttering the label's words. A large-vocabulary system allows users to switch between playing back recorded audio and speech recognition with a single input, with successive audio playbacks automatically starting slightly before the end of prior playback. And a cell phone that allows both large-vocabulary speech recognition and audio recording and playback.

Abstract: The error rate of a pronunciation guesser that guesses the phonetic spelling of words used in speech recognition is improved by causing its training to weigh letter-to-phoneme mappings used as data in such training as a function of the frequency of the words in which such mappings occur. Preferably the ratio of the weight to word frequency increases as word frequencies decreases. Acoustic phoneme models for use in speech recognition with phonetic spellings generated by a pronunciation guesser that makes errors are trained against word models whose phonetic spellings have been generated by a pronunciation guesser that makes similar errors. As a result, the acoustic models represent blends of phoneme sounds that reflect the spelling errors made by the pronunciation guessers. Speech recognition enabled systems are made by storing in them both a pronunciation guesser and a corresponding set of such blended acoustic models.

Abstract: The invention relates to the combination of speech recognition with handwriting and/or character recognition. This includes the innovation of selecting one or more best-scoring recognition candidates as a function of recognition of both handwritten and spoken representations of a sequence of one or more words to be recognized. It also includes the innovation of using character or handwriting recognition of one or more letters to alphabetically filter speech recognition of one or more words. It also includes the innovations of using speech recognition of one or more letter-identifying words to alphabetically filter handwriting recognition, and of using speech recognition to correct handwriting recognition of one or more words.

Abstract: The present invention relates to speech recognition that enables a user to perform re-utterance recognition, in which speech recognition is performed upon both a second saying of a sequence of one or more words and upon an earlier saying of the same sequence to help the speech recognition better select one or more best scoring text sequences for the utterances.

Abstract: The present invention relates to speech recognition using selectable recognition modes. This includes innovations such as: large vocabulary speech recognition programming that supplies recognized words to external program as they are recognized, and allows a user to select between large vocabulary recognition of an utterance with and without language context from the prior utterance independently of state of the external program; allowing a user to select between continuous and discrete speech recognition that use substantially the same vocabulary; allowing a user to select between continuous and discrete large-vocabulary speech recognition modes; allowing a user to select between at least two different alphabetic entry speech recognition modes; and allowing a user to select from among four or more of the following recognitions modes when creating text: a large-vocabulary mode, an alphabetic entry mode, a number entry mode, and a punctuation entry mode.

Abstract: The invention relates to methods, systems, and programming for displaying media in a scaled manner and/or in an orientation other than an operating system displays its graphical user interface. Some embodiments lay out digital content at a virtual pixel resolution and then display a portion of the layout at a smaller display resolution by displaying images and text at scaled-down coordinates and scaled-down sizes, with the display of text being composed from a plurality of font bitmaps having character shapes and pixel alignments selected to improve readability at the scaled-down size. Some embodiments, which operate on computing systems having operating systems that display a graphical user interface in a portrait orientation, display a scaled-down representation of a web page in a landscape orientation, including displaying in such a landscape orientation scaled-down images and scaled-down text composed from a plurality of font bitmaps designed for the scaled-down size.

Abstract: The present invention relates to: speech recognition using selectable recognition modes; using choice lists in large-vocabulary speech recognition; enabling users to select word transformations; speech recognition that automatically turns recognition off in one or more specified ways; phone key control of large-vocabulary speech recognition; speech recognition using phone key alphabetic filtering and spelling: speech recognition that enables a user to perform re-utterance recognition; the combination of speech recognition and text-to-speech (TTS) generation; the combination of speech recognition with handwriting and/or character recognition; and the combination of large-vocabulary speech recognition with audio recording and playback.

Abstract: A bitmap of a shape, such as a font, can be subpixel optimized by producing for each of a display's subpixels a coverage value representing the percent of its area covered by the shape being represented and by distributing, to prevent color imbalance, an amount of a given subpixel's coverage value to nearby subpixels of different colors as a function of the percent of the given subpixel's coverage value that causes color imbalance. Web pages can be displayed with scaled-down and subpixel optimized images. A given layout of a Web page can be displayed at each of at least two different selected scale factors, with the font bitmaps used to represent characters in the display at each scale factor having their shape and pixel alignment selected to improve readability for the particular pixel size at which they are displayed at each such scale factor.

Abstract: Web pages are displayed with a simultaneous overview and magnified view. An indicator can show the portion of the overview in the magnified view. Both views can be shown, one above the other, across the full width of the same screen. A user can select between such a split view and another view, including an overview-only view, a magnified-only view, and a view in which selected text is laid out to fit the width of the magnified view. Navigational input can directly move the layout in the magnified view or the cursor, and can scroll both the overview and magnified view. The magnified view can display text with antialiased fonts designed for its resolution. The magnified view can be made to function like a magnifying glass. The width of text in multicolumn layouts can be limited to fit the width of a view window, such as the magnified-view.

Abstract: A new word model is trained from synthetic word samples derived by Monte Carlo techniques from one or more prior word models. The prior word model can be a phonetic word model and the new word model can be a non-phonetic, whole-word, word model. The prior word model can be trained from data that has undergone a first channel normalization and the synthesized word samples from which the new word model is trained can undergo a different channel normalization similar to that to be used in a given speech recognition context. The prior word model can have a first model structure and the new word model can have a second, different, model structure. These differences in model structure can include, for example, differences of model topology; differences of model complexity; and differences in the type of basis function used in a description of such probability distributions.

Abstract: A computerized system, such as one practiced with an e-commerce font server, enables a user to select one or more fonts. According to one aspect, a user can select a font from a multi-dimensional font space defined by the attribute values of a plurality of fonts. The system repeatedly 1) enables the user to select a spatial specification defined relative to the currently selected font in the space; 2) determines which fonts, if any, match the spatial specification; 3) displays character-font shapes of those fonts; and 4) makes one of the displayed fonts a new currently selected font relative to which a new spatial specification can be selected. The spatial specification can merely selects fonts which are closest in the attribute space, can select the nearest font in a selected direction, or can provide more complex spatial selection. According to another aspect of the invention, a server makes fonts available to clients over a network.

Abstract: A computerized system lays out document templates represented as a tree of text and shape elements, including variable elements. The user can define a shape element to have a maximize or minimize property in one or more dimensions. The layout makes the minimized dimensions of a shape as small as its contents will allow; and makes the maximized dimensions expand as much as available space allows. Such maximization or minimization can be performed within a horizontal or vertical sequence box. Variable values mapped into variable shape elements can include sub-trees of text and/or shape elements, including shape elements which have the maximize or minimize property, and elements which are themselves variable elements. An anchor point can be fixed at a selected point on a shape, causing the anchor point to remain fixed as the rest of the shape expands or contract. Variable Image elements can maintain the aspect ratios of images mapped into them as those images are scaled.

Abstract: An apparatus and method for decreasing image acquisition and reconstruction times in magnetic resonance imaging are provided. Magnetic resonance data is acquired in parallel by an array of separate RF receiver coils disposed at generally circumferentially-spaced locations relative to one another around the imaging volume defined by the body coil of a magnetic resonance imaging apparatus. Further, the image reconstruction also is performed in parallel, thereby shortening up image display times. The apparatus and method operate on the basis of determining an estimate of the sensitivity profile of each RF coil in the array, and thereafter, utilizing those profiles in the creation of a desired image by encoding later acquired magnetic resonance data from a body of interest disposed in the imaging volume of the magnetic resonance imaging apparatus.

Abstract: A client-server system includes a new front end server which interposes itself between a prior (back end) server and its clients using an interposed dynamically-loaded library linked to the back end. The interposed library and front end can provide enhanced services without requiring reconfiguration of either the back end server or its clients. The front end can accept network connections on the same port as the back end server is programmed to. The interposed library intercepts standard network system calls by the back end and emulates their behavior by communicating with the front end. The front end can send file descriptors of either direct connections with clients, or of pipes for relaying requests and response from and to clients through the front end, to the interposed library for return to calls by the back end to accept connections.

Abstract: A Web system includes a server which receives URL requests from client computers and responds by transmitting the requested data to the clients along with performance monitoring instructions. These instructions cause the clients to send performance message back to the server indicating the time required on the client for certain acts associated with transmitted data. Often they cause a client to monitor the time required to download a composite data object, such as a Web page which include separately downloaded images, frames, or applets. They can also instruct the client to measure other times, such as the time required to download individual component objects; to execute given code, or to execute a transaction. Preferably the server includes means for automatically analyzing the performance messages it receives to detect different types of performance problems.

Abstract: A system rapidly rasterizes high resolution shapes, such as outline fonts, for use in a lower resolution pixel image. For individual pixels a line coverage value is determined for each of at least two sampling lines running in different directions, such as at right angles, within the pixel. The pixel's line coverage value for each line is a function of the degree to which the line is covered by any shapes within the pixel. Then a coverage value is determined for the pixel, itself, as a non-linear function of it's one or more line coverage values running in each of the different directions. Commonly the non-linear function causes the pixel's coverage value to vary more rapidly with variations in the line coverage value of that one of its sampling line which is closest to being half covered within the pixel.

Abstract: A computer interface allows as user to control aspects of its behavior by head motion. In one example, a head mounted display (HMD) includes a directional detector for generating an object detection signal when pointed at certain physical objects. The detector is mounted so users can point it by moving their heads. In another example, a graphical user interface includes the computation of a view space comprised of multiple separate portions, in each of which graphical user interface windows generated by separate computer processes can be selectively located by a user. The user can selectively move such windows between view space portions by a point and click interface. The method displays a subset of view space portions to the user at one time, senses motion of the user's head, and changes the portions of the view space displayed, in discrete units of view space portions, in response to changes in the position of the user's head.

Abstract: An optical disc protector includes a circular main surface and a peripheral lip. The main surface is a circular sheet of optically clear material having a diameter slightly larger than most optical discs on which it is to be mounted and having a mounting hole in its center. The peripheral lip is attached to, and extends all the way around, the perimeter of the circular main surface. It extends away from the main surface at an inward angle, so the smallest diameter of the peripheral lip is smaller than the diameter of the optical discs on which the protector is to be used. The peripheral lip is flexible so as to push against the circular edge of an optical disc upon which the protector is mounted. In many embodiments, the lip does not contain a clasp for grabbing against the side of the optical disc opposite from the side against which the main surface of the protector lies when mounted.

Abstract: A computer system provides an improved graphical user interface by defining a view space in which graphical user interface windows from separate computer processes can be selectively located by a user. This view space has a plurality of discrete portions each large enough to hold a plurality of such user interface windows. The system enables users to selectively associate view space portions with physical locations external to the system's computing apparatus. The system also enables users to selectively locate a variable number of the user interface windows in each of the view space portions associated with such a physical location, including enabling the user to selectively move windows between such view space portions.