749results about How to "Reduce ambiguity" patented technology

A speech processing system which exploits statistical modeling and formal logic to receive and process speech input, which may represent data to be received, such as dictation, or commands to be processed by an operating system, application or process. A command dictionary and dynamic grammars are used in processing speech input to identify, disambiguate and extract commands. The logical processing scheme ensures that putative commands are complete and unambiguous before processing. Context sensitivity may be employed to differentiate data and commands. A multi faceted graphic user interface may be provided for interaction with a user to speech enable interaction with applications and processes that do not necessarily have native support for speech input.

A human-computer interface system and methods for providing intelligent, adaptive, multimodal interaction with users while accomplishing tasks on their behalf in some particular domain or combination of domains. Specifically, this system accepts user input via natural language text, mouse actions, human speech, whistles, gestures, pedal movements, facial or postural changes, and conveys results via natural language text, automatically-generated speech, and displays of graphs, tables, animation, video, and mechanical and chemical effectors that convey heat, tactile sensation, taste and smell.

A system and method for processing speech input comprising a speech recognizer and a logical command processor which facilitates additional processing of speech input beyond the speech recognizer level. A speech recognizer receives input from a user, and when a command is identified in the speech input, if the command meets conditions that require additional processing, a representation of the speech input s stored for subsequent processing. A logical command processor performs additional processing of command input by analyzing the command and its elements, determining which elements are required for successful processing the command and which elements are present and lacking. The user is prompted to supply missing information, and subsequent user input is added to the command structure until the command input is aborted or the command structure reaches sufficient completeness to enable execution of the command. Thereby, speech input of complex commands in natural language in a system running a plurality of applications and processes is made possible.

In embodiments of the present invention, a system and computer-implemented method for enabling dictation may include parsing standard reports in order to identify a plurality of logical phrases in the report used for discrete sections and descriptions. In the report method, the phrases may be parsed and identifier words throughout the report may be compared to eliminate ambiguities. The method may then involve constructing text macros that follow the parsed text, thereby enabling the user to speak the identifiers to indicate full, formatted text. Finally, the report method may involve constructing a mnemonic document so both beginner and experienced users can easily read the identifiers out loud to produce a report. The result of the method is an intuitive, notes-style way to use speech commands to quickly produce a standard, formatted report.

Traffic measurement should make it possible to obtain the spatial flow of traffic through the domain, i.e., the paths or trajectories followed by packets between any ingress and egress point of the domain. A method of sampling packet trajectories in a packet switching network allows the direct inference of traffic flows through a measurement domain by observing the trajectories of a subset of all packets traversing the network. A method which assumes that the measurement domain does not change comprises the steps of selecting packets for sampling in accordance with a sampling function of the packet content and generating a practically unique label for each sampled packet. The method does not rely on routing state, its implementation cost is small, and the measurement reporting traffic is modest and can be controlled precisely. Using the same hash function will yield the same sample set of packets in the entire domain, and enables us to reconstruct packet trajectories. An alternate embodiment which assumes no constraints and that the measurement domain may change comprises the steps of applying a sampling function and altering an invariant bit position as a signaling flag in each packet selected for sampling.

Apparatus and methods are described for selecting which of a plurality of simultaneously activated keys in a keyboard based on capacitive sensors is a key intended for selection by a user. In embodiments of the invention keys are preferentially selected as the user intended key based on their positions within the keyboard. Thus a key which is frequently wrongly activated when a user selects another key, e.g. because the key is adjacent the intended key and the user normally passes his finger over it while approaching the desired key, can be suppressed relative to the desired key based on their relative positions. For example, keys may be associated with predefined rankings according to their position within the keyboard and in use keys are preferentially select according to their rankings. Alternatively, signals from the keys may be scaled by weighting factors associated with their positions and a key selected according to the weighted signals.

The present invention relates generally to the field of video-camera systems, such as a video conferencing systems, and more particularly to video camera targeting systems that locate and acquire targets using an input characterizing a target and a machine-classification system to assist in target acquisition responsively to that input. In some embodiments, the characterization and classification are employed together with one or more inputs of other modalities such as gesture-control. In one example of the system in operation, an operator is able to make pointing gestures toward an object and, simultaneously speak a sentence identifying the object to which the speaker is pointing. At least one term of the sentence, presumably, is associated with a machine-sensible characteristic by which the object can be identified. The system captures and processes the voice and gesture inputs and re-positions a PTZ video camera to focus on the object that best matches both the characteristics and the gesture. Thus, the PTZ camera is aimed based upon the inputs the system receives and the system's ability to locate the target by its sensors.

In a cellular mobile telecommunications system the position of a mobile station can be estimated in terms of its bearing and range from a cell site. A multi-element direction finding antenna at the cell site receives signals from the mobile station and a receiver circuit estimates the bearing using the relative phase of signals received at different antenna elements and estimates the range by measuring round trip delay of signals to and from the mobile station. Motion of the mobile station can introduce errors into the bearing estimate due to frequency offset and frequency spread when element sampling is non-simultaneous. Compensation for these errors is introduced by using signal samples successively received at the same antenna element to estimate Doppler frequency offset and spread. It is necessary to ensure accurate calibration of the direction finding antenna and the receiver circuit. This is done by injecting calibration signals into the circuit near the antenna or into the antenna itself from a near field probe. Other aspects of calibration, such as antenna position, are calibrated using a remote beacon. A beacon emulating a mobile station but at a fixed, known location, or a beacon at an adjacent cell site may be used.

The present invention relates to a navigation unit and base station used for determining location. A plurality of base stations are initialized to determine their location relative to each other. At the navigation unit, the time of arrival of at least one signal from each of the plurality of base stations is measured. From this, the location of the navigation unit relative to the plurality of base stations may be directly calculated using a closed solution. In one embodiment, a time of arrival technique is used and in another embodiment a time difference of arrival technique is used. Preferably an ultra-wide band frequency is utilized.

The present invention discloses a visual target detection and labeling method. The method includes: an image inputting step, to input an image to be detected; a candidate region extracting step, to extract a candidate window as the candidate region from the image to be detected using selectively search algorithm; a feature description extracting step, to perform feature description on the candidate region using a pre-trained large-scale convolutional neural network and output the feature description of the candidate region; a visual target predicting step, to predict the candidate region based on the feature description of the candidate region using a pre-trained object detection module, to estimate regions having the visual target; and a position labeling step, to labeling the position of the visual target according to the estimated result. Experiments show that, compared with the mainstream week supervision visual target detection and labeling method, the present invention has a stronger ability to excavate positive samples and a more general application prospect, and is suitable for visual target detection and automatic labeling tasks on the large-scale data set.

A system and method that translates sentences of natural language text into sets of axioms of formal logic that are consistent with parses resulting from NLP and acquired constraints as they accumulate. The system and method further present these axioms so as to facilitate further disambiguation of such sentences and produces axioms of formal logic suitable for processing by automated reasoning technologies, such as first-order or description logic suitable for processing by various reasoning algorithms, such as logic programs, inference engines, theorem provers, and rule-based systems.

A virtual environment for user interaction, and virtual objects for use therein, the virtual objects having an internal co-ordinate system with natural language definition to allow simplified inter-object relationships. The internal co-ordinate system is used to define docking positions for connecting to other objects to allow logical or natural positional relationships. A dynamically defined menu system allows pop-up menus for simple user entry of required relationships.

A method and system for examining tissue in order to differentiate it from other tissue according to the dielectric properties of the examined tissue, by: applying a probe to the tissue to be examined, such that the probe generates an electrical fringe field in the zone of the examined tissue and produces a reflected pulse therefrom with negligible radiation penetrating into the tissue itself; detecting the reflected electrical pulse; and comparing electrical characteristics of the reflected electrical pulse with respect to the applied electrical pulse to provide an indication of the dielectric properties of the examined tissue.

The invention discloses an intelligent electromechanical equipment management system and method based on BIM. The intelligent electromechanical equipment management system comprises an integrated delivery platform, an equipment information management module, a maintenance management module, an operation and maintenance knowledge base module and an emergence plan management module. By virtue of the system and the method, a property management practitioner can efficiently realize management of equipment information, maintenance, an operation and maintenance knowledge base and an emergence plan of a large number of complex electromechanical equipment of a building.

An improved method of estimating an earth model utilizes an acoustic Full Waveform Inversion (FWI) of seismic data in a pseudo-time coordinate system, in which the earth model m({tilde over (m)}) is parameterized with two lateral coordinates (x,y) and a vertical coordinate ({tilde over (z)}) which expresses vertical travel time of acoustic signals used to generate the seismic data.

Systems and methods for creating and using a domain profile include identifying a status of a first page associated with a domain. The first page is retrieved and additional pages from the domain are identified based on hyperlinks from the first page. The status of the additional pages is identified and the hyperlinks are prioritized based on the status and/or a comparison with predetermined data. Content is extracted from the first page and selected pages from among the additional pages. The specific additional pages may be selected based on the prioritization. The retrieved content may be processed through a signature marker set to determine a contextual match. A purpose of the domain is determined according to the status of the first page, the status of the additional pages and results of the processing of the content. The domain profile can be displayed, stored, sent and/or searched to identify web sites or attributes of interest.