Microsoft Technology Licensing, LLC

Other

Bo-June Hsu

Kuansan Wang

Yu-ting Kuo

Heung-Yeung Shum

Jose Alberto Rodriguez

Peter Gregory Davis

Eric Norman Badger

Chao-Chia Liu

Lynn Dai

Raymond Quan

Dan Hwang

Chiyuan Huang

Jeremy Espenshade

Alejandro Acero

Timothy S Paek

Robert L Chambers

Bongshin Lee

Milind V Mahajan

Arun Sacheti

Emre Mehmet Kiciman

Philipp Heinz Schmid

Sanaz Ahari

Nitin Agrawal

Xiaolong Li

Farid Hosseini

Robert A Williams

Chun-Kai Wang

Maxim V Stepin

Richard Tunnicliffe

Lawrence William Colagiovanni

Cornelia Carapcea

David Jeremy Guy Wood

Jason A Bolla

Yusuf Furkan Fidan

Oscar H Newkerk, Iii

Described herein are technologies that are configured to assist a user in recollection information about people, places, and things. Computer-readable data is captured, and contextual data that temporally corresponds to the computer-readable data is also captured. In a database, the computer-readable data is indexed by the contextual data. Thus, when a query is received that references the contextual data, the computer-readable data is retrieved.

Extended memory system

The description relates to predicting text based on multiple user attempts at inputting text. One example can include a computing device comprising a user interface. In this example, the user interface is configured to receive evidence from a first user attempt at inputting text and configured to receive evidence from a second user attempt at inputting the text. The computing device also includes a text prediction engine configured to combine the evidence from the first user attempt at inputting the text and the evidence from the second user attempt at inputting the text to predict the text.

Predicting text by combining attempts

Example apparatus and methods concern controlling a hover-sensitive input/output interface. One example apparatus includes a proximity detector that detects an object in a hover-space associated with the input/output interface. The apparatus produces characterization data concerning the object. The characterization data may be independent of where in the hover-space the object is located. The apparatus selectively controls the activation, display, and deactivation of user interface elements displayed by the apparatus on the input/output interface as a function of the characterization data and interface state. Selectively controlling the activation, display, and deactivation of the user interface elements includes allocating display space on the input/output interface to the user interface elements when they are needed for an operation on the apparatus and selectively reclaiming space on the input/output interface allocated to the user interface elements when they are not needed for an operation on the apparatus.

Hover controlled user interface element

A dictation module is described herein which receives and interprets a complete utterance of the user in incremental fashion, that is, one incremental portion at a time. The dictation module also provides rendered text in incremental fashion. The rendered text corresponds to the dictation module's interpretation of each incremental portion. The dictation module also allows the user to modify any part of the rendered text, as it becomes available. In one case, for instance, the dictation module provides a marking menu which includes multiple options by which a user can modify a selected part of the rendered text. The dictation module also uses the rendered text (as modified or unmodified by the user using the marking menu) to adjust one or more models used by the dictation model to interpret the user's utterance.

Dictation with incremental recognition of speech

Technologies pertaining to providing completions to proffered prefixes are disclosed herein. A suggested completion to a proffered prefix is retrieved by walking nodes of a trie data structure, wherein a node includes one or more characters that are used to extend a character sequence represented by its parent. Each node in the trie data structure is assigned a score, wherein the score maps to a best score assigned to its descendants. The nodes of the trie data structure are sorted based upon score, and the nodes are walked based upon scores assigned thereto.

Retrieval of prefix completions by way of walking nodes of a trie data structure

Systems, methods and computer-storage media are provided for identifying query formulation suggestions in response to receiving a search query. A portion of a search query is received. Query formulation suggestions are identified by semantically analyzing the search query. The query formulation suggestions are used to further formulate the received search query. The query formulation suggestions include semantic-pattern-based query suggestions that are derived from semantic query patterns, one or more entities, and information associated with these entities. The query formulation suggestions are transmitted for presentation.

Interactive semantic query suggestion for content search

Systems, methods and computer-storage media are provided for identifying low-match search queries and determining comparable item matches to suggest to the user in response to a low-match query. 'Low-match queries' are queries for which an insufficient number of exact item matches are available. In embodiments, exact and/or comparable item matches may be determined via semantic analysis. Also provided are systems, methods and computer-storage media for informing the user, by way of a presented indicator, or the like, that a presented item was selected for presentation based upon a similarity metric rather than being determined an exact match for the input query.

Identifying query formulation suggestions for low-match queries

The present invention is directed to presenting a suggested search query. Responsive to receiving a user-devised search parameter, a suggested search query is identified. The user-devised search parameter might have been previously received by a search system, or alternatively, might be a unique query that has not been previously received. A suggested search query might be generated using various techniques, such as by applying an n-gram language model. A classification of the suggested search query is determined, and the suggested search query is presented together with a visual indicator, which signifies the classification.

Generating and presenting a suggested search query

Architecture that employs a modeling technique based on language modeling to estimate a probability of a document matching the user need as expressed in the query. The modeling technique is based on the data mining results that various portions of a document (e.g., body, title, URL, anchor text, user queries) use different styles of human languages. Thus, the results based on a language can be adapted individually to match the language of query. Since the approach is based on adaptation, the framework also provides a natural means to progressively revise the model as user data are collected. Different styles of languages in a document can be recognized and adapted individually. Background language models are also employed that offer a fallback approach in case the document has incomplete fields of data, and can utilize topical or semantic hierarchy of the knowledge domain.

Bootstrap and adapt a document search engine

A resource selection system is described for assisting a user in performing a task that includes multiple actions. At each stage of the task, the system presents a set resources from which the user may select to perform a subsequent action in the task. The system implicitly selects the set of resources based on context information that identifies the user's current informational needs. For example, the context information may be derived from textual information that is being presented on a user device, which the user is presumed to be viewing at the current time. In one implementation, the system selects the set of resources by computing language models for respective domains and respective entities. The system uses the language models to determine the relevance of the context information to each of the domains. The system then selects resources associated with domains that have been assessed as relevant.

Federated implicit search

A method of entering information into a mobile device includes receiving a multi-word speech input from a user, performing speech recognition on the speech input to obtain a multi-word speech recognition result, and sequentially displaying, in a display, words in the speech recognition result for user confirmation or correction, by adding one word at a time to the display. A next word is only displayed after user confirmation or correct has been received for a previously displayed word that is immediately preceding the next word in the speech recognition result. The method also includes calculating a hypothesis lattice indicative of a plurality of speech recognition hypotheses based on the speech input and, prior to finishing calculating the hypothesis lattice and while continuing to calculate the hypothesis lattice, calculating a preliminary hypothesis lattice indicative of only partial speech recognition hypotheses based on the speech input and outputting the preliminary hypotheses lattice.

Combined speech and alternate input modality to a mobile device

Speech recognition utilizing categories and prefixes is disclosed. Categories identify types of recognition and allow different grammars and prefixes for each category. Categories can be directed to specific applications and/or program modules. Uttering a prefix allows users to easily direct text to specific grammars for enhanced recognition, and also to direct the recognized text to the appropriate application/module.

Speech recognition using categories and speech prefixing

Growing community of inventors

Woodinville, WA, United States of America

Bo-June Hsu