Google LLC patent applications on December 26th, 2024
Patent Applications by Google LLC on December 26th, 2024
Google LLC: 45 patent applications
Google LLC has applied for patents in the areas of G06N20/00 (4), G10L15/22 (4), G06F40/40 (4), G06V20/40 (3), G10L15/197 (3) G10L15/22 (3), H04W76/40 (2), G06F40/40 (2), G06F40/35 (2), G06V20/41 (2)
With keywords such as: user, device, content, data, based, computing, video, methods, determining, and input in patent application abstracts.
Patent Applications by Google LLC
Inventor(s): Christopher David Workman of Baltimore MD (US) for google llc, Vahe Tshitoyan of Fremont CA (US) for google llc, Altaf Khetani of Denver CO (US) for google llc, Cheongyuen William Tsang of Union City CA (US) for google llc, Kelly Elizabeth Dobson of Mountain View CA (US) for google llc
IPC Code(s): A61B5/145, A61B5/1455, G01N21/65, G01N33/66
CPC Code(s): A61B5/14532
Abstract: the present disclosure provides computer-implemented methods, systems, and devices for non-invasively measuring a level of an analyte in a user using stimulated raman scattering. a user computing device includes a raman pump source that emits pump light toward a skin surface of the user at a pump wavelength. the device includes a stokes source that emits stokes light toward the skin surface at a plurality of stokes wavelengths within a window of raman measurement wavelengths. the stokes light comprises a plurality of narrowband emissions at a respective plurality of center wavelengths ranging across the window of raman measurement wavelengths. the device includes a photodetector that measures light that emanates from the skin surface. the device includes a processor that processes the measured light to provide an estimated analyte level of the user.
Inventor(s): Catherine Hsiao of San Francisco CA (US) for google llc, Alan Merzon of San Jose CA (US) for google llc, Kristie Fisher of Venice CA (US) for google llc, Stacey Peterson of Los Gatos CA (US) for google llc, Haochen Liu of San Jose CA (US) for google llc, Nicholas Dunn of Mountain View CA (US) for google llc, Ramachandra Shivputrappa of Mountain View CA (US) for google llc, Sam Cheung of San Francisco CA (US) for google llc, Erica Hansen of Los Altos CA (US) for google llc, Ben Frenkel of Santa Clara CA (US) for google llc
IPC Code(s): A63F13/795, A63F13/355, A63F13/48, A63F13/86
CPC Code(s): A63F13/795
Abstract: methods, systems, and media for coordinating multiplayer game sessions are provided. in some embodiments, the method comprises: determining that a first user associated with a first user device is viewing game content created by a second user associated with a second user device; in response to determining that the first user is viewing the game content created by the second user, determining whether the first user satisfies criteria for joining a multiplayer game session with the second user; in response to determining that the first user satisfies criteria for joining the multiplayer game session with the second user, causing a selectable input element to join a queue of players to participate in the multiplayer game session with the second user to be presented on the first user device; in response to determining that the selectable input element to join the queue of players has been selected on the first user device, adding the first user to a position of the queue of players; in response to determining that the first user can participate in the multiplayer games session based on the position within the queue of players, causing an invitation to join the multiplayer game session to be presented on the first user device; in response to determining that the invitation to join the multiplayer game session has been accepted on the first user device, causing the first user to be added to the multiplayer game session; causing the multiplayer game session to begin on the first user device, wherein the multiplayer game session includes at least the first user and the second user; and in response to determining that the multiplayer game session has ended, removing the first user from the multiplayer game session.
Inventor(s): Valey Kamalov of Gainesville FL (US) for google llc., Matthew Eldred Newland of San Jose CA (US) for google llc., Massimiliano Salsi of Sunnyvale CA (US) for google llc., Mattia Cantono of San Clara CA (US) for google llc., Vijayanand Vusirikala of Palo Alto CA (US) for google llc.
IPC Code(s): G01V1/01, G01H9/00, G01V1/28, G01V1/30
CPC Code(s): G01V1/01
Abstract: systems and methods are provided for generating a model for detection of seismic events. in this regard, one or more processors may receive from one or more stations located along an underwater optical route, one or more time series of polarization states of a detected light signal during a time period. the one or more processors may transform the one or more time series of polarization states into one or more spectrums in a frequency domain. seismic activity data for the time period may be received by the one or more processors, where the seismic activity data include one or more seismic events detected in a region at least partially overlapping the underwater optical route. the one or more processors then generate a model for detecting seismic events based on the one or more spectrums and the seismic activity data.
20240427075. WAVEGUIDES INCLUDING A BLUE WAVEGUIDE FILM_simplified_abstract_(google llc)
Inventor(s): Eliezer Glik of San Diego CA (US) for google llc, Pierre St. Hilaire of Belmont CA (US) for google llc
IPC Code(s): F21V8/00, G02B27/00, G02B27/01
CPC Code(s): G02B6/0076
Abstract: a stacked waveguide is configured to provide light representative of an image to the eye of a user. to this end, the stacked waveguide includes a first waveguide having a first refraction index and including an incoupler configured to direct a first portion of display light having a first wavelength and a second portion of display light that has a second, different wavelength into the first waveguide. further, the stacked waveguide includes a second waveguide having a second, different refraction index and including an incoupler configured to direct a third portion of display light having a third wavelength different from the first and second wavelengths into the second waveguide.
20240427150. CURVED THIN SEE-THROUGH LIGHTGUIDE WITH LARGE EYEBOX_simplified_abstract_(google llc)
Inventor(s): Ozan Cakmakci of Sunnyvale CA (US) for google llc
IPC Code(s): G02B27/01, G02B1/10, G02B27/28
CPC Code(s): G02B27/0172
Abstract: a display system includes a microdisplay and a curved light-guide to direct and magnify light from the microdisplay to a user's eye. the display system includes an incoupler that is thicker than the curved lightguide to couple light from the microdisplay into the curved lightguide and an out-coupler tilted at an angle to receive two bounces of display light and couple the display light out of the curved lightguide, resulting in a large eyebox. in some embodiments, the curved lightguide includes a low-index coating on a world-side surface with a gap through which side image light escapes and a sunglass coating on a lens to absorb the side image light. the curved light-guide includes a half-wave plate (hwp) to flip an input polarization for a main image to an orthogonal polarization and a dielectric mirror to act as a weak analyzer to dim side images.
Inventor(s): Lu Tian of Palo Alto CA (US) for google llc, Wei Jin of Saratoga CA (US) for google llc, Kang Luo of San Jose CA (US) for google llc, Thomas Mercier of Weston FL (US) for google llc
IPC Code(s): G03F7/00, B29C59/00, B29C59/02, G02B6/13
CPC Code(s): G03F7/0002
Abstract: a method of nanoimprinting a waveguide includes forming a working stamp having at least one waveguide feature pattern and at least one process control feature pattern. the working stamp is pressed into a waveguide workpiece thereby forming waveguide features in one or more functional waveguide zones of the waveguide workpiece and one or more sets of process control features in one or more regions outside of the one or more functional waveguide zones. the working stamp is detached from the waveguide workpiece.
20240427395. POWER SEQUENCER FOR POWER STATE MANAGEMENT_simplified_abstract_(google llc)
Inventor(s): Ronak Subhas Patel of Sunnyvale CA (US) for google llc, William James Tuohy of Danville CA (US) for google llc, Vinu Vijay Kumar of Belmont CA (US) for google llc, Alex Robert Carlson of Sunnyvale CA (US) for google llc
IPC Code(s): G06F1/26
CPC Code(s): G06F1/266
Abstract: methods, systems, and apparatus for handling applications in an ambient computing system. one of the apparatus includes multiple devices arranged in multiple power blocks, wherein each device of the multiple devices belongs to one of the multiple power blocks; and multiple local power managers, each local power manager being programmable to execute respective sets of instruction sequences for a respective power block in order to effectuate power state transitions for one or more devices in the respective power block.
Inventor(s): Dmytro Ievdokymenko of Rancho Santa Margarita CA (US) for google llc
IPC Code(s): G06F3/0482, G06F9/451, G06F16/28, H04L51/42
CPC Code(s): G06F3/0482
Abstract: implementations generate adaptive labels for a variety of different documents in furtherance of re-organizing the documents and/or otherwise simplifying a presentation of at least a subset of the documents at an application interface. a label that is generated for a given document can be based on content of the document, which can be processed using one or more different artificial intelligence techniques to determine the label. a set of labels generated for the variety of different documents can be utilized for organizing a subset of the documents that are associated with a particular user. in these and other manners, when the user accesses an application, the user can more readily sift through their documents in a more computationally efficient manner, thereby preserving computational resources being consumed to present the documents to the user. for example, as opposed to a user being presented with the same static labels for filtering and/or sorting over time, the labels can adapt over time and adapt in accordance with the contents of underlying emails or other documents.
20240427479. Video Conference Virtual Whiteboard_simplified_abstract_(google llc)
Inventor(s): Arthur Edmond Blume of Beverly MA (US) for google llc
IPC Code(s): G06F3/04842, G06F3/0484, G06V10/764, G06V20/70, H04L65/403
CPC Code(s): G06F3/04842
Abstract: a computing system and method that can be used for a video conferencing system including a graphical user interface having a virtual whiteboard user interface that is responsive to inputs from multiple attendees of a video conference. in particular, the state of an application tool such as a camera, visual note indicator, web link, pen/pencil tool, highlighter tool, or other application tool executing or otherwise represented within the multi-user interaction whiteboard can respond and change in real-time as multiple attendees of the video conference interact with the virtual whiteboard user interface. the systems and methods of the present disclosure allow for a multi-user interaction whiteboard to offer various tools for the video conferencing system to facilitate interactions by multiple attendees within the graphical user interface.
Inventor(s): Yan Huang of Mountain View CA (US) for google llc, Nikhil Rao of Santa Clara CA (US) for google llc, Owen Lin of Mountain View CA (US) for google llc, Ilya Firman of Sunnyvale CA (US) for google llc, Anil Iyer of Sunnyvale CA (US) for google llc
IPC Code(s): G06F8/60, G06F9/54, G06F11/36, G06F16/9035, G06F16/906, G06F16/907
CPC Code(s): G06F8/60
Abstract: analyzing or debugging applications is provided. the system identifies an action for an application provided by a developer. the system determines a first classification score based on historical execution of the action. the system generates a machine generated action for the application based on metadata associated with the application. the system determines a second classification score based on a comparison of the action with the machine generated action. the system selects, via a matching program, a second application that matches the application. the system determines a third classification score based on a comparison of an action approved for the second application with the action provided by the application developer. the system updates a delivery control parameter based on the first classification score, the second classification score and the third classification score. the system controls delivery of the application based on the delivery control parameter.
20240427702. Memory Copy in Mesh Networks_simplified_abstract_(google llc)
Inventor(s): Roi Yosef Leibowitch Ben Ari of Jerusalem (IL) for google llc
IPC Code(s): G06F12/0817, G06F12/0811
CPC Code(s): G06F12/0822
Abstract: aspects of the disclosure are directed to configuring one or more system level caches to perform memory copy in a mesh network. the system level caches can perform a write to a write destination or perform a write to another system level cache that performs the write to the write destination. the system level caches allow for avoiding redundant traffic, as data no longer needs to be sent to and from a cpu when performing memory copy.
20240427770. QUERY RESTARTABILITY_simplified_abstract_(google llc)
Inventor(s): Yevgeniy Kogan of Kirkland WA (US) for google llc, Rajesh Rao of Clyde Hill WA (US) for google llc, Sergey Melnik of Kirkland WA (US) for google llc
IPC Code(s): G06F16/2455, G06F16/245, G06F16/2453
CPC Code(s): G06F16/2455
Abstract: methods, systems, and apparatus, including computer programs encoded on computer storage media, for restarting a query using a token. one of the methods includes receiving, by a computer from a requesting device, a query; determining, using a data storage system, a current result responsive to the query; generating, using the current result, a restart token that represents operations performed to determine a plurality of results responsive to the query including the current result responsive to the query and that can be used to determine a new result responsive to the query that was not included in the plurality of results responsive to the query; and providing, to the requesting device, a message that includes a) first data for the restart token that represents operations performed to determine the plurality of results responsive to the query and b) second data for the current result responsive to the query.
Inventor(s): Vikram Aggarwal of Palo Alto CA (US) for google llc, Alice Lee of Menlo Park CA (US) for google llc
IPC Code(s): G06F16/2457, G06F16/29, G06F16/335, G06F16/9535, G06F16/9537
CPC Code(s): G06F16/24575
Abstract: user attribute data associated with a user may be used to resolve unresolved entities in connection with the performance of computer-based actions, e.g., actions resulting from action queries generated based at least in part on content being accessed on a computing device.
20240427923. SECURED MANAGEMENT OF DATA DISTRIBUTION RESTRICTIONS_simplified_abstract_(google llc)
Inventor(s): Gang Wang of Frederick MD (US) for google llc, Kevin Wei Li Yeo of New York NY (US) for google llc
IPC Code(s): G06F21/62
CPC Code(s): G06F21/6218
Abstract: methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for allowing suitable digital components to be automatically selected and provided to a client device. methods can include generating a universal identifier for a digital component that is presented in the application. the application updates a set of universal identifiers that has been created for digital components presented by the application over a specified time period. the application identifies digital components and the corresponding universal identifiers that are blocked and generates a probabilistic data structure representing the set of blocked universal identifiers. the application creates multiple shares of the probabilistic data structure and transmits different shares to different servers. the application receives a separate response generated by each of the different servers based on the multiple shares and identifies a digital component to present in the application based on a combination of the separate responses.
20240427931. PRIVACY PRESERVING DATA COLLECTION AND ANALYSIS_simplified_abstract_(google llc)
Inventor(s): Gang Wang of Frederick MD (US) for google llc, Marcel M. Moti Yung of New York NY (US) for google llc, Andres Munoz Medina of Brooklyn NY (US) for google llc
IPC Code(s): G06F21/62, G06F16/9535, H04L9/32
CPC Code(s): G06F21/6245
Abstract: a method includes receiving, by a data processing apparatus and from a content distribution system, a message comprising a probabilistic data structure representing a set of content items that should not be provided to a user device, content item data for content items available to be provided, and a request to determine whether any content item data is invalid, determining that the content item data for a given content item is invalid because the given content item may be in the set of content items represented by the probabilistic data structure, including removing the content item data for the given content item that was determined to be invalid; and preventing distribution of content items including the given content item.
Inventor(s): Andrei Tudor Stratan of San Jose CA (US) for google llc, Olivier Jean Benoit of San Diego CA (US) for google llc
IPC Code(s): G06F21/75, G06F21/31, G06F21/57, G06F21/60, G06F21/79
CPC Code(s): G06F21/75
Abstract: this document describes methods and systems that use an undefined lifecycle state identifier to manage security of a system-on-chip (soc) integrated circuit (ic) device. as part of the described techniques, the soc ic device may include a first set of logic integrated circuitry that determines that a first combination of bit values fails to correspond to a known lifecycle state identifier. the first set of logic integrated circuitry may then provide, to a second set of logic integrated circuitry, a second combination of bit values that corresponds to the undefined lifecycle state identifier. the second set of logic integrated circuitry may then place the soc ic device into an undefined lifecycle state.
Inventor(s): Khalid Salama of London (GB) for google llc, Ágoston Weisz of Pfaeffikon (CH) for google llc
IPC Code(s): G06F40/30, G06F40/40
CPC Code(s): G06F40/30
Abstract: a method includes obtaining a set of training queries that each specify a corresponding operation to perform and include a corresponding plurality of speech recognition hypotheses that each represent a corresponding candidate transcription of the training query, and a corresponding ground-truth transcription of the training query. for each training query, the method includes processing, using an encoder of a neural semantic parsing (nsp) model, the corresponding plurality of speech recognition hypotheses to generate a corresponding nsp embedding, processing, using a transcription decoder, the corresponding nsp embedding to generate a corresponding predicted transcription, and determining a corresponding first loss based on the corresponding predicted transcription and the corresponding ground-truth transcription. the method further includes training, based on the first losses determined for the set of training queries, the nsp model to learn how to predict user intents associated with the operations specified by the training queries.
Inventor(s): Swarna Srinivas Kakodkar of Bethlehem PA (US) for google llc, Arpita Saha of Mountain View CA (US) for google llc
IPC Code(s): G06F40/35, G06Q30/015
CPC Code(s): G06F40/35
Abstract: implementations described herein relate to methods, devices, and computer-readable media to automatically modify a content item based on computer-mediated interaction with reference to a content item. the method includes programmatically analyzing the computer-mediated interaction between a user and a customer service agent to determine if the computer-mediated interaction is successful. the method further includes identifying information provided by the customer service agent during the computer-mediated interaction. the method further includes generating a content update based on the identified information. the method further includes modifying the content item based on the content update.
Inventor(s): Sarvjeet Singh of Palo Alto CA (US) for google llc, Gustavo Menezes Ponte Moreira of Sunnyvale CA (US) for google llc, Grady Simon of San Francisco CA (US) for google llc, Peter Brandt of Palo Alto CA (US) for google llc, Tushar Chandra of Los Altos CA (US) for google llc
IPC Code(s): G06F40/35, G06F40/40, H04L51/02
CPC Code(s): G06F40/35
Abstract: techniques are disclosed that enable the generation of a content agent based on content parameter(s) determined from an initial user request for content as well as a dialog session to further refine the request for content. various implementations include using the content agent to render additional content responsive to an additional user request. additional or alternatively implementations include using the content agent to proactively render content to the user.
20240428006. LARGE LANGUAGE MODEL (LLM) QUANTIZATION_simplified_abstract_(google llc)
Inventor(s): Jian Li of Mountain View CA (US) for google llc, Zhifeng Chen of Sunnyvale CA (US) for google llc, Yanping Huang of Mountain View CA (US) for google llc, Yuanzhong Xu of Mountain View CA (US) for google llc, Tao Wang of Sunnyvale CA (US) for google llc, YaGuang Li of Sunnyvale CA (US) for google llc
IPC Code(s): G06F40/40
CPC Code(s): G06F40/40
Abstract: implementations relate to asymmetric quantization of large language models (llms). processor(s) of a system can: obtain a trained llm, wherein the trained llm includes a plurality of layers, each layer comprising a respective plurality of weights; for each layer of the plurality of layers: calculate an optimal clipping range for the respective plurality of weights, and clip one or more weights of the respective plurality of weights that lie outside of the optimal clipping range to produce a clipped layer; quantize the llm to generate a quantized llm, wherein the instructions to quantize include instructions to map weights of the plurality of clipped layers of the llm from continuous values to discrete values; and provide the quantized llm for downstream processing.
Inventor(s): Jinsung Yoon of San Jose CA (US) for google llc, Jiefeng Chen of Madison WI (US) for google llc, Sayna Ebrahimi of Los Altos Hills CA (US) for google llc, Sercan Omer Arik of San Francisco CA (US) for google llc
IPC Code(s): G06F40/40
CPC Code(s): G06F40/40
Abstract: aspects of the disclosure are directed to methods, systems, and computer readable media for adaptation with self-evaluation to improve selective prediction in large language models (llms), generally referred to as aspire. aspire includes training llms on a portion of training data from a question answering task to learn self-evaluation, e.g., learn to distinguish whether a generated answer is correct or not. aspire further includes a selection score that combines a likelihood of that generated answer is correct with a self-evaluation score for selective prediction. aspire demonstrates improved selective prediction performance with less computational cost.
20240428056. PERFORMING TASKS USING GENERATIVE NEURAL NETWORKS_simplified_abstract_(google llc)
Inventor(s): Paul Kishan Rubenstein of Zurich (CH) for google llc, Matthew Sharifi of Kilchberg (CH) for google llc, Alexandru Tudor of Uitikon (CH) for google llc, Chulayuth Asawaroengchai of Zurich (CH) for google llc, Duc Dung Nguyen of Zurich (CH) for google llc, Marco Tagliasacchi of Ruvigliana (CH) for google llc, Neil Zeghidour of Paris (FR) for google llc, Zalán Borsos of Zurich (CH) for google llc, Christian Frank of Zurich (CH) for google llc, Dalia Salem Hassan Fahmy Elbadawy of Zurich (CH) for google llc, Hannah Raphaelle Muckenhirn of Zurich (CH) for google llc, Dirk Ryan Padfield of Seattle WA (US) for google llc, Damien Vincent of Zurich (CH) for google llc, Evgeny Kharitonov of Paris (FR) for google llc, Michelle Dana Tadmor of Tel Aviv (IL) for google llc, Mihajlo Velimirovic of Zurich (CH) for google llc, Feifan Chen of Seattle WA (US) for google llc, Victoria Zayats of Seattle WA (US) for google llc
IPC Code(s): G06N3/0475, G10L25/30
CPC Code(s): G06N3/0475
Abstract: methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing tasks. one of the methods includes obtaining a sequence of input tokens, where each token is selected from a vocabulary of tokens that includes text tokens and audio tokens, and wherein the sequence of input tokens includes tokens that describe a task to be performed and data for performing the task; generating a sequence of embeddings by embedding each token in the sequence of input tokens in an embedding space; and processing the sequence of embeddings using a language model neural network to generate a sequence of output tokens for the task, where each token is selected from the vocabulary.
Inventor(s): David Richard So of Brooklyn NY (US) for google llc, Quoc V. Le of Sunnyvale CA (US) for google llc, Hanxiao Liu of Santa Clara CA (US) for google llc, Wojciech Andrzej Manke of Zürich (CH) for google llc, Zihang Dai of Pittsburgh PA (US) for google llc, Noam M. Shazeer of Palo Alto CA (US) for google llc
IPC Code(s): G06N3/08
CPC Code(s): G06N3/08
Abstract: methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing a machine learning task on a network input to generate a network output. one of the systems includes an attention neural network configured to perform the machine learning task. the attention neural network includes one or more attentions layers that each include a squared relu activation layer, a depth-wise convolution layer, or both.
20240428137. Typicality of Batches for Machine Learning_simplified_abstract_(google llc)
Inventor(s): Elad Edwin Tzvi Eban of Sunnyvale CA (US) for google llc, Alan Mackey of Mountain View CA (US) for google llc, Piotr Zielinski of New Providence NJ (US) for google llc
IPC Code(s): G06N20/00
CPC Code(s): G06N20/00
Abstract: systems and methods described herein can improve typicality of batches for machine learning. the systems and methods can include obtaining a corpus of training data, the corpus of training data including one or more training examples. the systems and methods can include generating a first batch set including a plurality of batches from the corpus of training data, each of the batches including a subset of the one or more training examples. the systems and methods can include determining a batch distribution of a first batch of the first batch set. the systems and methods can include determining that the first batch is an atypical batch based on the batch distribution of the first batch. the systems and methods can include, in response to determining that the first batch is an atypical batch, shuffling the training examples of the first batch and one or more second batches of the first batch set to generate a second batch set. the systems and methods can include training a first machine-learned model using the second batch set.
20240428473. Real Screens in Extended Reality_simplified_abstract_(google llc)
Inventor(s): Diego Rivas Vetencourt of San Francisco CA (US) for google llc, Yilong Yao of San Jose CA (US) for google llc, David Kim of Zug (CH) for google llc, Steven Soon Leong Toh of Alameda CA (US) for google llc
IPC Code(s): G06T11/00, G06F3/01, G06K7/14, G06T7/70, G06V20/20, G06V20/40, H04L67/131
CPC Code(s): G06T11/00
Abstract: methods and systems for displaying screens in extended reality are disclosed herein. the method can include receiving video data of a real-world environment of a user of an extended reality device and identifying a screen of a separate computing device in the video data. the method can also include pairing the extended reality device and the separate computing device, generating a graphical outline around the screen of the separate computing device, and displaying, by the processor, the screen of the separate computing device and the graphical outline around the screen of the separate computing device on a display of the extended reality device. the method can further include detecting a user interaction with the graphical outline around the screen of the separate computing device based on the received video data and in response to detecting the user interaction, performing an action associated with the user interaction.
Inventor(s): Joseph Lemberg of Menlo Park CA (US) for google llc
IPC Code(s): G06T13/80
CPC Code(s): G06T13/80
Abstract: a multimedia communication system for enabling an end user computing device to render a moving image on a display device. the system has an image render command generator that is arranged to provide an image rendering command that is executable by a web browser of the end user computing device to cause the end user computing device to render the moving image on a frame-by-frame basis from a still image in multimedia content by slicing the still image into a plurality of rows or columns of pixels and modifying a subset of the plurality of rows or columns of pixels between frames, and a multimedia content packager arranged to assemble a multimedia content package having the multimedia content, including the still image and the image rendering command.
Inventor(s): Tanmay Shah of San Ramon CA (US) for google llc, Vishal Vinod of San Diego CA (US) for google llc, Dmitry Lagun of San Jose CA (US) for google llc
IPC Code(s): G06T15/04, G06T7/70, G06T17/00
CPC Code(s): G06T15/04
Abstract: provided are systems and methods for creating 3d representations from one or more images of objects. it involves training a machine-learned correspondence network to convert 3d locations of pixels into a 2d canonical coordinate space. this network can map texture values from ground truth or synthetic images of the object into the 2d space, creating a texture data set. when a new synthetic image is generated from a specific pose, the 3d locations can be mapped into the 2d space, allowing texture values to be retrieved and applied to the new image. the system also enables users to edit the texture data, facilitating texture edits and transfers across objects.
Inventor(s): Ariel Fuxman of Redwood City CA (US) for google llc, Alexander Kenji Hata of Mountain View CA (US) for google llc, Edward Benjamin Vendrow of Redwood City CA (US) for google llc, Otilia Stretcu of Fremont CA (US) for google llc, Wenlei Zhou of Sunnyvale CA (US) for google llc, Krishnamurthy Viswanathan of Sunnyvale CA (US) for google llc, Aditya Avinash of Santa Clara CA (US) for google llc, Gabriel Berger of Cupertino CA (US) for google llc, Andrew Ames Bunner of Belmont CA (US) for google llc, Javier Alejandro Rey of San Mateo CA (US) for google llc, Wei Qiao of Santa Clara CA (US) for google llc, Yintao Liu of Union City CA (US) for google llc, Guanzhong Wang of Los Angeles CA (US) for google llc, Thomas Nathan Denby of Santa Monica CA (US) for google llc, Mehmet Nejat Tek of Redwood City CA (US) for google llc, Neil Gordon Alldrin of Campbell CA (US) for google llc, Enming Luo of Mountain View CA (US) for google llc, Chun-Ta Lu of Sunnyvale CA (US) for google llc
IPC Code(s): G06V10/778, G06V10/764, G06V10/774, G06V10/82, G06V10/94
CPC Code(s): G06V10/7788
Abstract: a computer-implemented method includes receiving an input from a user relating to a concept, automatically obtaining a first set of images from an unlabeled dataset of images based on the input, and obtaining a first rating via the user for each image from the first set of images. the method further includes training a classifier model relating to the concept based on the first set of images rated by the user, automatically obtaining a second set of images from the unlabeled dataset of images based on the classifier model trained based on the first set of images, and obtaining a second rating via the user for each image from the second set of images. the classifier model relating to the concept is retrained based on the first set of images rated by the user and the second set of images rated by the user to obtain an updated classifier model.
20240428586. Systems and Methods for Improved Video Understanding_simplified_abstract_(google llc)
Inventor(s): Anurag Arnab of Grenoble (FR) for google llc, Mostafa Dehghani of Amsterdam (NL) for google llc, Georg Heigold of Aachen (DE) for google llc, Chen Sun of San Francisco CA (US) for google llc, Mario Lucic of Adliswil (CH) for google llc, Cordelia Luise Schmid of Saint-Ismier (FR) for google llc
IPC Code(s): G06V20/40, G06N20/00
CPC Code(s): G06V20/41
Abstract: a computer-implemented method for classifying video data with improved accuracy includes obtaining, by a computing system comprising one or more computing devices, video data comprising a plurality of video frames; extracting, by the computing system, a plurality of spatiotemporal representations from the video data, the plurality of spatiotemporal representations comprising a representation of spatiotemporal information in the video data; providing, by the computing system, the plurality of spatiotemporal representations as input to a video understanding model, the video understanding model comprising a video transformer encoder model; and receiving, by the computing system, a classification output from the video understanding model.
20240428587. Systems and Methods for Improved Video Understanding_simplified_abstract_(google llc)
Inventor(s): Anurag Arnab of Grenoble (FR) for google llc, Mostafa Dehghani of Amsterdam (NL) for google llc, Georg Heigold of Aachen (DE) for google llc, Chen Sun of San Francisco CA (US) for google llc, Mario Lucic of Adliswil (CH) for google llc, Cordelia Luise Schmid of Saint-Ismier (FR) for google llc
IPC Code(s): G06V20/40, G06N20/00
CPC Code(s): G06V20/41
Abstract: a computer-implemented method for classifying video data with improved accuracy includes obtaining, by a computing system comprising one or more computing devices, video data comprising a plurality of video frames; extracting, by the computing system, a plurality of video tokens from the video data, the plurality of video tokens comprising a representation of spatiotemporal information in the video data; providing, by the computing system, the plurality of video tokens as input to a video understanding model, the video understanding model comprising a video transformer encoder model; and receiving, by the computing system, a classification output from the video understanding model.
Inventor(s): Petar Aleksic of Jersey City NJ (US) for google llc, Pedro J. Moreno Mengibar of Jersey City NJ (US) for google llc, Leonid Velikovich of New York NY (US) for google llc
IPC Code(s): G10L15/197, G10L15/16, G10L15/18, G10L15/187
CPC Code(s): G10L15/197
Abstract: methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing contextual grammar selection are disclosed. in one aspect, a method includes the actions of receiving audio data of an utterance. the actions include generating a word lattice that includes multiple candidate transcriptions of the utterance and that includes transcription confidence scores. the actions include determining a context of the computing device. the actions include based on the context of the computing device, identifying grammars that correspond to the multiple candidate transcriptions. the actions include determining, for each of the multiple candidate transcriptions, grammar confidence scores that reflect a likelihood that a respective grammar is a match for a respective candidate transcription. the actions include selecting, from among the candidate transcriptions, a candidate transcription. the actions further include providing, for output, the selected candidate transcription as a transcription of the utterance.
Inventor(s): Ke Hu of Stony Brook NY (US) for google llc, Tara N. Sainath of Jersey City NJ (US) for google llc, Arun Narayanan of Milpitas CA (US) for google llc, Ruoming Pang of New York NY (US) for google llc, Trevor Strohman of Mountain View CA (US) for google llc
IPC Code(s): G10L15/197, G06F40/126, G10L15/02, G10L15/06, G10L15/08, G10L15/22
CPC Code(s): G10L15/197
Abstract: a method includes receiving a sequence of acoustic frames and generating, by a first encoder, a first higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. the method also includes generating, by a first pass transducer decoder, a first pass speech recognition hypothesis for a corresponding first higher order feature representation and generating, by a text encoder, a text encoding for a corresponding first pass speech recognition hypothesis. the method also includes generating, by a second encoder, a second higher order feature representation for a corresponding first higher order feature representation. the method also includes generating, by a second pass transducer decoder, a second pass speech recognition hypothesis using a corresponding second higher order feature representation and a corresponding text encoding.
20240428790. DETERMINING DIALOG STATES FOR LANGUAGE MODELS_simplified_abstract_(google llc)
Inventor(s): Petar Aleksic of Jersey City NJ (US) for google llc, Pedro J. Moreno Mengibar of Jersey City NJ (US) for google llc
IPC Code(s): G10L15/22, G06F40/295, G06F40/30, G10L15/065, G10L15/183, G10L15/197, G10L15/26
CPC Code(s): G10L15/22
Abstract: systems, methods, devices, and other techniques are described herein for determining dialog states that correspond to voice inputs and for biasing a language model based on the determined dialog states. in some implementations, a method includes receiving, at a computing system, audio data that indicates a voice input and determining a particular dialog state, from among a plurality of dialog states, which corresponds to the voice input. a set of n-grams can be identified that are associated with the particular dialog state that corresponds to the voice input. in response to identifying the set of n-grams that are associated with the particular dialog state that corresponds to the voice input, a language model can be biased by adjusting probability scores that the language model indicates for n-grams in the set of n-grams. the voice input can be transcribed using the adjusted language model.
Inventor(s): Ulas Kirazci of Mountain View CA (US) for google llc, Adam Coimbra of Los Altos CA (US) for google llc, Abraham Lee of Belmont CA (US) for google llc, Wei Dong of Union City CA (US) for google llc, Thushan Amarasiriwardena of Alameda CA (US) for google llc
IPC Code(s): G10L15/22, G06F3/16, G06F9/448, G10L13/027
CPC Code(s): G10L15/22
Abstract: techniques are described herein for multi-modal interaction between users, automated assistants, and other computing services. in various implementations, a user may engage with the automated assistant in order to further engage with a third party computing service. in some implementations, the user may advance through dialog state machines associated with third party computing service using both verbal input modalities and input modalities other than verbal modalities, such as visual/tactile modalities.
Inventor(s): Raunaq Shah of San Francisco CA (US) for google llc, Jaclyn Konzelmann of Mountain View CA (US) for google llc, Lisa Takehana of San Bruno CA (US) for google llc, Ruxandra Davies of Santa Monica CA (US) for google llc, Adrian Diaconu of Mountain View CA (US) for google llc
IPC Code(s): G10L15/22, G06F3/16
CPC Code(s): G10L15/22
Abstract: implementations set forth herein relate to employing dynamic regulations for governing responsiveness of multiple automated assistant devices, and specifically the responsiveness an automated assistant to a given spoken utterance that has been acknowledged by two or more of the assistant devices. the dynamic regulations can be context-dependent and adapted over time in order that the automated assistant can accommodate assistant interaction preferences that may vary from user to user. for instance, a spoken utterance such as “stop,” may be intended to affect different assistant actions based on a context in which the user provided the spoken utterance. the context can refer to a location of the user relative to other rooms in a home, a time of day, a user providing the spoken utterance, an arrangement of the assistant devices within a home, and/or a state of each device in the home.
20240428816. AUDIO-VISUAL HEARING AID_simplified_abstract_(google llc)
Inventor(s): Anatoly Efros of Rishon LeTsiyon (IL) for google llc, Noam Etzion-Rosenberg of Binyamina (IL) for google llc, Tal Remez of Ramat Gan (IL) for google llc, Oran Lang of Ramat Gan (IL) for google llc, Inbar Mosseri of Raanana (IL) for google llc, Israel Or Weinstein of Tel Aviv (IL) for google llc, Benjamin Schlesinger of Ramat Hasharon (IL) for google llc, Michael Rubinstein of Natick MA (US) for google llc, Ariel Ephrat of Efrat (IL) for google llc, Yukun Zhu of Shoreline WA (US) for google llc, Stella Laurenzo of Seattle WA (US) for google llc, Amit Pitaru of Brooklyn NY (US) for google llc, Yossi Matias of Tel Aviv (IL) for google llc
IPC Code(s): G10L21/0208, G10L17/00, G10L21/0272, G10L25/57
CPC Code(s): G10L21/0208
Abstract: methods, systems, and apparatus, including computer programs encoded on computer storage media, for audio-visual speech separation. a method includes: receiving, by a user device, a first indication of one or more first speakers visible in a current view recorded by a camera of the user device, in response, generating a respective isolated speech signal for each of the one or more first speakers that isolates speech of the first speaker in the current view and sending the isolated speech signals for each of the one or more first speakers to a listening device operatively coupled to the user device, receiving, by the user device, a second indication of one or more second speakers visible in the current view recorded by the camera of the user device, and in response generating and sending a respective isolated speech signal for each of the one or more second speakers to the listening device.
20240428818. AUDIO-FOCUS FOR AMBIENT NOISE CANCELLATION_simplified_abstract_(google llc)
Inventor(s): Rajeev Nongpiur of Mountain View CA (US) for google llc, Neil Zeghidour of Paris (FR) for google llc, Marco Tagliasacchi of Ruvigliana (CH) for google llc
IPC Code(s): G10L21/0364, G10L15/06, G10L21/0208, G10L21/0232, G10L21/034, G10L25/30
CPC Code(s): G10L21/0364
Abstract: a method including identifying an audio capture device and a target direction associated with the audio capture device, detecting first audio associated with the target direction, enhancing the first audio using a machine learning model configured to detect audio associated with the target direction, optionally, detecting second audio associated with a direction different from the target direction, and optionally, diminishing the second audio using the machine learning model.
20240428937. Instruction Prompt Tuning for Machine-Learned Models_simplified_abstract_(google llc)
Inventor(s): Vivek Natarajan of Sunnyvale CA (US) for google llc, Karan Singhal of Roslyn NY (US) for google llc, Shekoofeh Azizi of Vancouver (CA) for google llc, Alan Prasana Karthikesalingam of London (GB) for google llc, Tao Tu of New York NY (US) for google llc, Seyedeh Sara Mahdavi of Oakville (CA) for google llc, Christopher Semturs of San Mateo CA (US) for google llc
IPC Code(s): G16H50/20
CPC Code(s): G16H50/20
Abstract: an aspect of the present disclosure provides an example method comprising: receiving an input query associated with a particular task domain of a plurality of available task domains; obtaining a machine-learned prompt component and a curated prompt component, wherein the machine-learned prompt component comprises a plurality of machine-learned prompt values for the plurality of available task domains, and wherein the curated prompt component comprises a plurality of exemplar prompt values corresponding to one or more embedded natural language exemplars for the particular task domain from domain experts; and generating an output responsive to the input query by processing a combined prompt and the input query using a pre-trained machine-learned model, wherein the combined prompt comprises the machine-learned prompt component and the curated prompt component.
20240429866. Qubit Control Electronics_simplified_abstract_(google llc)
Inventor(s): Joseph Cheney BARDIN of Amherst MA (US) for google llc
IPC Code(s): H03D7/00, B82Y10/00, G06F1/20, G06N10/00, H01F6/00, H02M1/08, H03D7/14, H03K3/38, H03K17/693, H03K17/92, H03K19/195, H03M1/12, H04B10/077, H04B10/70
CPC Code(s): H03D7/005
Abstract: a device for generating a qubit control signal includes: a first signal envelope generator circuit including a first multiple of signal sources, in which an output of each signal source of the first multiple of signal sources is combined to provide a first cumulative output; and a first mixer circuit coupled to the first signal envelope generator circuit, in which the first cumulative output is coupled to a first input of the first mixer circuit, and an output of the first mixer circuit includes a first qubit control signal.
Inventor(s): Pooja Mittal of Palo Alto CA (US) for google llc, Kumara Guruparan Tharmalingam of Los Gatos CA (US) for google llc
IPC Code(s): H04L9/32
CPC Code(s): H04L9/3239
Abstract: example embodiments of the present disclosure provide for an example method for on-device location history for content analytics. the example method includes obtaining data associated with a first user and a second user. the example method includes generating content event groupings associated with the users. the example method includes a user device requesting the content event groupings. the example method includes obtaining de-identified content-location event pairs generated on the user device. the example method includes inputting the de-identified content-location event pairs into an attribution pipeline. the example method includes obtaining attribution data associated with conversions attributed to specific content-location event pairs.
20240430415. Adaptive Filter Computation Precision in Video Coding_simplified_abstract_(google llc)
Inventor(s): Yunqing Wang of Palo Alto CA (US) for google llc
IPC Code(s): H04N19/117, H04N19/176, H04N19/80
CPC Code(s): H04N19/117
Abstract: filtering for a block of pixels that uses variable rounding bits to change the resolution of a filter result is described. the rounding bits are derived adaptively for the block for high precision in the filtering calculations. while useful to change the bit depth in single-stage filtering, the techniques are particularly useful for multi-stage filtering. a first filtering operation may be performed on the block having an input bit depth. the output is an intermediate filter result with a precision greater than the input bit depth. a rounding bit value for modifying the precision of the intermediate filter result is adaptively determined. the precision is modified using the rounding bit value. thereafter, a second filtering operation may be performed on the intermediate filter result. an output of the second filtering operation is a filtered block of pixels, with which a coding operation may be performed.
20240430567. Depth-Histogram Based Autofocus_simplified_abstract_(google llc)
Inventor(s): Tongyang Liu of () for google llc, Leung Chun Chan of Sunnyvale CA (US) for google llc, Ying Chen Lou of Santa Clara CA (US) for google llc, Hsuan Ming Liu of Taipei (TW) for google llc, Han-Lei Wang of New Taipei City (TW) for google llc
IPC Code(s): H04N23/67, G06T7/50
CPC Code(s): H04N23/672
Abstract: a method includes receiving a plurality of depth values corresponding to a plurality of areas depicted in an image captured by an image sensor. the method also includes generating a depth histogram categorizing each depth value of the plurality of depth values into a depth value range of a plurality of depth value ranges. the method further includes determining an autofocus distance based on the depth histogram. the method additionally includes causing the image sensor to capture an image based on the autofocus distance.
20240430577. LOW-LIGHT AUTOFOCUS TECHNIQUE_simplified_abstract_(google llc)
Inventor(s): Ying Chen Lou of Mountain View CA (US) for google llc, Leung Chun Chan of Mountain View CA (US) for google llc, Kiran Murthy of Mountain View CA (US) for google llc, Qiurui He of Mountain View CA (US) for google llc, Szepo Robert Hung of Mountain View CA (US) for google llc, Sushil Nath of Mountain View CA (US) for google llc
IPC Code(s): H04N23/73, H04N23/71
CPC Code(s): H04N23/73
Abstract: the present disclosure relates to a low-light autofocus technique. one example embodiment includes a method. the method includes receiving an indication of a low-light condition for a camera system. the method also includes determining an extended exposure time for a low-light autofocus procedure of the camera system. further, the method includes capturing, by the camera system, an extended frame for the low-light autofocus procedure. the extended frame is captured by the camera system using the determined extended exposure time. in addition, the method includes determining, based on the captured extended frame, an in-focus lens setting for a lens of the camera system.
20240430632. Real-time Adaptation of Audio Playback_simplified_abstract_(google llc)
Inventor(s): Ricardo Antonio Garcia of Mountain View CA (US) for google llc, Gustav David Gunnar Sennton of London (GB) for google llc, Bich Ha Lo of London (GB) for google llc, Nicholas James Clark of London (GB) for google llc, Nicole Laure McWilliams of London (GB) for google llc, Richard Ledley of London (GB) for google llc, Alex Florescu of Hampton (GB) for google llc, Jamie Garside of London (GB) for google llc
IPC Code(s): H04R29/00, G06N20/00, H03G5/16
CPC Code(s): H04R29/001
Abstract: a method includes determining one or more audio characteristics of a first digital signal representing a first portion of streaming media content provided by a media playback application to an audio output component; playing the first portion of the media content into an environment; capturing, by an audio output component, a second digital signal representing the first portion of the media content as played; determining, based on the captured second digital signal, one or more detected audio characteristics of the media content as played into the environment; determining a difference between the one or more audio characteristics of the media content and the one or more detected audio characteristics; calibrating a second portion of the media content based on the difference and based on a determination that the second portion is to be calibrated; and playing, by the audio output component into the environment, the second portion as calibrated.
Inventor(s): Chih-Hsiang Wu of Taoyuan City (CN) for google llc, Teming Chen of Taoyuan City (CN) for google llc
IPC Code(s): H04W76/40, H04W76/27
CPC Code(s): H04W76/40
Abstract: in a method of managing multicast and/or broadcast services (mbs) communications after a state transition, a ue receives mbs data from a ran using mbs configuration parameters while the ue is in a connected state with the ran. the ce transitions from the connected state to an idle state or an inactive state. while the ue is in the idle state or the inactive state, and based on whether the mbs configuration parameters are for receiving a broadcast mbs session or a non-broadcast mbs session, the ue either continues or does not continue, respectively, to receive mbs data using the mbs configuration parameters.
Inventor(s): Chih-Hsiang Wu of Taoyuan City (TW) for google llc
IPC Code(s): H04W76/40, H04W72/1263
CPC Code(s): H04W76/40
Abstract: in a method for configuring a radio bearer, a distributed unit (du) of a distributed base station receives, from a central unit (cu) of the distributed base station, a request to configure radio resources for a radio bearer and, based on one or more factors associated with the radio bearer, either includes or refrains from including in a response one or more configuration parameters for the radio bearer. the du also transmits the response to the cu.
- Google LLC
- A61B5/145
- A61B5/1455
- G01N21/65
- G01N33/66
- CPC A61B5/14532
- Google llc
- A63F13/795
- A63F13/355
- A63F13/48
- A63F13/86
- CPC A63F13/795
- G01V1/01
- G01H9/00
- G01V1/28
- G01V1/30
- CPC G01V1/01
- Google llc.
- F21V8/00
- G02B27/00
- G02B27/01
- CPC G02B6/0076
- G02B1/10
- G02B27/28
- CPC G02B27/0172
- G03F7/00
- B29C59/00
- B29C59/02
- G02B6/13
- CPC G03F7/0002
- G06F1/26
- CPC G06F1/266
- G06F3/0482
- G06F9/451
- G06F16/28
- H04L51/42
- CPC G06F3/0482
- G06F3/04842
- G06F3/0484
- G06V10/764
- G06V20/70
- H04L65/403
- CPC G06F3/04842
- G06F8/60
- G06F9/54
- G06F11/36
- G06F16/9035
- G06F16/906
- G06F16/907
- CPC G06F8/60
- G06F12/0817
- G06F12/0811
- CPC G06F12/0822
- G06F16/2455
- G06F16/245
- G06F16/2453
- CPC G06F16/2455
- G06F16/2457
- G06F16/29
- G06F16/335
- G06F16/9535
- G06F16/9537
- CPC G06F16/24575
- G06F21/62
- CPC G06F21/6218
- H04L9/32
- CPC G06F21/6245
- G06F21/75
- G06F21/31
- G06F21/57
- G06F21/60
- G06F21/79
- CPC G06F21/75
- G06F40/30
- G06F40/40
- CPC G06F40/30
- G06F40/35
- G06Q30/015
- CPC G06F40/35
- H04L51/02
- CPC G06F40/40
- G06N3/0475
- G10L25/30
- CPC G06N3/0475
- G06N3/08
- CPC G06N3/08
- G06N20/00
- CPC G06N20/00
- G06T11/00
- G06F3/01
- G06K7/14
- G06T7/70
- G06V20/20
- G06V20/40
- H04L67/131
- CPC G06T11/00
- G06T13/80
- CPC G06T13/80
- G06T15/04
- G06T17/00
- CPC G06T15/04
- G06V10/778
- G06V10/774
- G06V10/82
- G06V10/94
- CPC G06V10/7788
- CPC G06V20/41
- G10L15/197
- G10L15/16
- G10L15/18
- G10L15/187
- CPC G10L15/197
- G06F40/126
- G10L15/02
- G10L15/06
- G10L15/08
- G10L15/22
- G06F40/295
- G10L15/065
- G10L15/183
- G10L15/26
- CPC G10L15/22
- G06F3/16
- G06F9/448
- G10L13/027
- G10L21/0208
- G10L17/00
- G10L21/0272
- G10L25/57
- CPC G10L21/0208
- G10L21/0364
- G10L21/0232
- G10L21/034
- CPC G10L21/0364
- G16H50/20
- CPC G16H50/20
- H03D7/00
- B82Y10/00
- G06F1/20
- G06N10/00
- H01F6/00
- H02M1/08
- H03D7/14
- H03K3/38
- H03K17/693
- H03K17/92
- H03K19/195
- H03M1/12
- H04B10/077
- H04B10/70
- CPC H03D7/005
- CPC H04L9/3239
- H04N19/117
- H04N19/176
- H04N19/80
- CPC H04N19/117
- H04N23/67
- G06T7/50
- CPC H04N23/672
- H04N23/73
- H04N23/71
- CPC H04N23/73
- H04R29/00
- H03G5/16
- CPC H04R29/001
- H04W76/40
- H04W76/27
- CPC H04W76/40
- H04W72/1263
