Microsoft Technology Licensing, LLC

Microsoft Corporation

Dachuan Zhang

Shawn Alan Villaron

Nathan Robert Penner

Sara Melanie Brown

Robert Parker

Ji Li

Amit Srivastava

Ryan Charles Hill

Jeffrey Huang

Siddharth Agrawal

Amit Kumar

Adit Krishnan

Michael John Arvold

Gilberto Aristides Apodaca Aragon

Mingxi Cheng

Sumithra Bhakthavatsalam

Christopher Robert Hayworth

Kushal Gurunath Sukthankar

Melissa Kristine Hunter

Rebecca Meredith Loew

Jonathan Patrick Schoeller

Andrew Hewatt

Greeshma Marri

Maya Bisineer

Andi Chen

Veena Gopalakrishna Joshi

A data processing system implements techniques for generating personalized content using a brand kit. The system receives a natural language prompt to generate content in a design application on the client device of a user and analyzes the prompt to determine whether the user intends to apply a brand kit to the generated content. The system automatically generates a brand kit for the user if one does not already exist and applies the brand kit to content generated using one or more generative models to create personalized content. The system includes a prompt generation unit that generates a plurality of model-specific prompts to the one or more generative models to cause the one or more generative models to create the personalized content.

Personalized branding with prompt adaptation in large language models and visual language models

A system and method and for retrieving assets from a personalized asset library includes receiving a search query for searching for assets in one or more asset libraries, the one or more asset libraries including a personalized asset library; encoding the search query into embedding representations via a trained query representation machine-learning (ML) model; comparing, via a matching unit, the query embedding representations to a plurality of asset representations, each of the plurality of asset representations being a representation of one of the plurality of candidate assets; identifying, based on the comparison, at least one of the plurality of the candidate assets as a search result for the search query; and providing the identified plurality of candidate assets for display as the search result. The plurality of asset representations for the one or more assets in the personalized content library are generated automatically without human labeling.

Method and system of retrieving assets from personalized asset libraries

Examples are disclosed that relate to methods and computing devices for providing voice-based assistance during a presentation. In one example, a method comprises receiving content of a slide deck, processing the content of the slide deck, and populating a contextual knowledge graph based on the content of the slide deck. A voice input is received from a presenter. Using the knowledge graph, the voice input is analyzed to determine an action to be performed by a presentation program during the presentation. The action is translated into one or more commands executable by the presentation program to perform the action, and the one or more commands are sent to a client device executing the presentation program.

Contextual voice-based presentation assistance

Systems, components, devices, and methods for transferring files are provided. An example file is a container file comprising a plurality of component files. A non-limiting example method is a method for transferring a container file over a network to a remote device. The method includes the step of accessing a container file, the container file comprising a plurality of components. The method also includes the step of transmitting metadata information for the plurality of components to the remote device. The method further includes the step of determining components from the plurality of components to transmit to the remote device. The method additionally includes the step of transmitting component data items for the determined components to the remote device.

Transferring files

In general, this disclosure describes techniques that employ user-defined values of properties of sections of an electronic presentation. As described herein, a user may configure a presentation to include a plurality of sections. Each of the sections includes zero or more slides of the electronic slide presentation. In addition to the slides associated with each section, each of the sections is associated with one or more properties having values that can be defined by an author of the presentation. Because the values of the properties of the sections are user-definable, these properties may, in some implementations of these techniques, enable the author of the presentation to use the sections in ways not possible in presentation applications that merely use sections as a means of grouping thumbnail images of slides for navigation among slides in an authoring interface.

Sections of a presentation having user-definable properties

Technologies are described herein for reducing network bandwidth utilization during file transfer. An application or another type of program identifies embedded objects in a file. The application then replaces the embedded objects with corresponding unique placeholder objects that are more highly compressible than the embedded objects. The application then compresses the file containing the unique placeholder objects. The application then transmits the compressed file to a network service that provides functionality for converting the file format of the file or for processing the file in another manner. The processed file generated by the network service also includes the unique placeholder objects. When the application receives the processed file from the network service, the application replaces the unique placeholder objects in the processed file with the embedded objects from the original file. The application can then open the processed file for viewing, editing, or another purpose.

Reducing network bandwidth utilization during file transfer

A command stream is generated that includes serialized commands for editing a document. The command stream can be applied to a modified document to generate a single document that contains modifications to the document made using both a Web application and a client application. The command stream can also be utilized to recreate the edited state of a document following the disconnection from a Web application for editing the document, to load balance a server computer hosting the Web application even while editing sessions are in progress, to perform an upgrade of a server hosting the Web application while editing sessions are in progress, and for other purposes.

Serializing document editing commands

Remote batch editing of formatted text may be provided. A request to edit a source document comprising text using a first markup language may be received. The source document may be converted to a second markup language and provided to the requestor. Changes to the converted document may be received and edit commands may be generated for those changes. The edit commands may then be applied to the source document.

Remote batch editing of formatted text via an HTML editor

Resolution and composition of electronic document layout are provided. An intermediate text data structure may be generated to hold a 'resolved' rich text state for a given document. Properties contained in the 'resolved' rich text state are a composite of all relevant properties including user defined and entered properties and including properties associated with the document according to a pre-built document context. This text body resolution process then may be utilized for generating a composite text layout for the text streams associated with a plurality of document components for generating a single rich text stream for presentation to and editing by a user.

Composition of electronic document layout

An edit list including a collection of change descriptions and definitions for a primary data structure is used for on-demand, incremental updating of associated auxiliary data structures. Edits performed on the primary data structure are recorded in the edit list with a version stamp, along with definitions of the changes and any other pertinent data. Each auxiliary data structure maintains its own version stamp that identifies when the auxiliary data structure was last synchronized with the primary data structure. When an auxiliary data structure needs to be updated, the edit list is retrieved and change descriptions scanned to find one with the matching version stamp. All change descriptions logged since that one are incorporated into the auxiliary data structure. The specific contents of each change description enable the auxiliary data structure to be incrementally updated as each change description is processed individually.

On-demand incremental update of data structures using edit list

Selecting and formatting warped text is disclosed. Text effects such as an underline effect and/or a strikethrough effect may be transformed along with rectangular text to form a warped text element. In such a situation, the text effects are warped with the characters of the text. Also, warped text may be accurately selected or 'hit.' The coordinates of a selection of warped text are inversely transformed to map the selection to the rectangular text. The inverse transform provides an accurate manner to determine the scope and location of the selection. When the scope and location are determined, the graphical representation of the selection may be transformed to provide a user with an accurate representation of a warped selection.

Selecting and formatting warped text

Methods, systems, and computer program products adapt and render a graphic effect associated with text rendered via an output device. A method involves receiving a character associated with a graphic effect and determining whether the graphic effect has not been fully rendered with a previously received character via the output device. Graphic effects that are fully rendered are rendered in full fidelity without degradation. In response to determining that the graphic effect has not been fully rendered with the previously received character, the method further involves degrading a display quality of the graphic effect and rendering the graphic effect with degradation and rendering the character to allow the character and the graphic effect to be displayed via the output device substantially in real time.

Adapting and rendering graphic effects

Highlighting selected objects in an electronic document is provided in a manner such that a highlighted object maintains coloring and formatting properties to allow for a maintenance of visual fidelity with a pre-highlighted view of the highlighted object and that allows for a semi-transparent background underneath a highlighted object that allows a user to see and appreciate the coloring and any visual properties applied to the background underneath the highlighting applied to the selected object.

Growing community of inventors

Sunnyvale, CA, United States of America

Dachuan Zhang