GOOGLE LLC patent applications on December 19th, 2024
Patent Applications by GOOGLE LLC on December 19th, 2024
GOOGLE LLC: 42 patent applications
GOOGLE LLC has applied for patents in the areas of G10L15/22 (11), G10L15/26 (6), G10L15/30 (4), G06F3/16 (4), G10L15/16 (4) G10L15/22 (4), H04W72/30 (3), G10L15/16 (2), G10L15/30 (1), G10L17/18 (1)
With keywords such as: device, data, user, input, based, speech, image, utterance, assistant, and network in patent application abstracts.
Patent Applications by GOOGLE LLC
20240418994. CURVED SEE-THROUGH LIGHTGUIDE WITH INTERMEDIATE FOCUS_simplified_abstract_(google llc)
Inventor(s): Eliezer Glik of San Diego CA (US) for google llc, Ozan Cakmakci of San Bruno CA (US) for google llc
IPC Code(s): G02B27/01, G02B6/42
CPC Code(s): G02B27/0172
Abstract: a head-mounted display includes a microdisplay configured to emit light representative of an image and provide the emitted light to an optics relay included in the head-mounted display. the optics relay is configured to magnify the image emitted by the microdisplay by a predetermined power such that an intermediate image based on the magnified image is formed in a curved lightguide of the head-mounted display. the intermediate image is then provided by the curved lightguide to an optical output system included in or otherwise coupled to the curved lightguide. the optical output system then provides at least a portion of the intermediate image to the eye of a user.
Inventor(s): Ammar Husain of San Francisco CA (US) for google llc, Ting Lu of Santa Clara CA (US) for google llc
IPC Code(s): G05D1/00
CPC Code(s): G05D1/0214
Abstract: a method includes receiving one or more past trajectories navigated by a robotic device in an environment, wherein the one or more past trajectories are associated with initial environmental sensor data and one or more obstacle detection heuristics. the method also includes determining, based at least on subsequent environmental sensor data, one or more updated obstacle detection heuristics. the method further includes determining, based on the one or more updated obstacle detection heuristics and the initial environmental sensor data, one or more predicted drivable areas in the environment. the method additionally includes, based on the one or more predicted drivable areas including the one or more past trajectories, using the one or more updated obstacle detection heuristics to determine future navigation of the robotic device.
Inventor(s): Dongeek Shin of San Jose CA (US) for google llc
IPC Code(s): G06F9/451, G06F3/044, G06F3/16
CPC Code(s): G06F9/453
Abstract: implementations set forth herein relate to controlling invocation of an automated assistant according to whether a capacitive touch sensor array has detected a particular input that indicates a user has positioned an assistant-enabled device near their face. the capacitive touch sensor array can be part of a touch display interface of a portable computing device that provides access to an automated assistant. when the interface is positioned near the face of the user, input data from the interface can be processed to determine whether the input data indicates the display interface is near their face or whether the user is providing some other input to the display interface. when the input data indicates the user is positioning the display interface near their face or mouth, the automated assistant can be invoked in lieu of the user providing any other invocation input.
Inventor(s): Christopher Thomas Kennelly of Croton-on-Hudson NY (US) for google llc, Xinliang David Li of Palo Alto CA (US) for google llc
IPC Code(s): G06F12/0806, G06F9/50, G06F9/54
CPC Code(s): G06F12/0806
Abstract: the disclosed technology provides instructions to manipulate the state of a processor's cache. an application may send an instruction directly to the processor, providing it with information about memory allocation or deallocation events occurring within the application. this direct communication facilitates immediate synchronization between the application and the processor, ensuring accurate memory management.
Inventor(s): Keun Soo Yim of San Jose CA (US) for google llc, Adam Coimbra of Cupertino CA (US) for google llc
IPC Code(s): G06F16/248, G06F16/2453, G06F16/2455
CPC Code(s): G06F16/248
Abstract: implementations include receiving search-based content donated by a first application installed at a client device, and processing the search-based content to generate a first entry in a central on-device repository that locally stores content donated by different applications installed at the client device. implementations further include receiving, via a unified interface that is independent of the first and second applications, a search query from the user and, in response to receiving the search query via the unified interface, searching the central on-device repository to determine whether any entry in the central on-device repository is responsive to the search query. if it is determined that the first entry is responsive to the search query, an interface element is generated based on the first entry, and the generated interface element is rendered at the unified interface, for potential selection by the user.
Inventor(s): Zijie Li of New York NY (US) for google llc, Shitanshu Verma of Livingston NJ (US) for google llc, Can Tang of Brooklyn NY (US) for google llc, Gary Elliott of Larchmont NY (US) for google llc, Gregory Allen Morris of Hanover NH (US) for google llc, Thomas Robert Magrino of Mamaroneck NY (US) for google llc, Jack Timothy Dingilian of Brooklyn NY (US) for google llc, Teng Zhong of Mountain View CA (US) for google llc, Andrii Shyshkalov of Munich (DE) for google llc, Siu Man Yau of Plainview NY (US) for google llc, Yijie Bu of Sunnyvale CA (US) for google llc
IPC Code(s): G06F16/27, G06F16/23, G06F16/25, G06F16/28
CPC Code(s): G06F16/273
Abstract: a method for change data capture state tracking includes accessing a distributed database including a plurality of clusters, each cluster of the plurality of clusters including a respective plurality of partitions, each cluster of the plurality of clusters configured to receive read and write operation. the method includes receiving, at a second cluster, a plurality of changes for a second table and storing the plurality of changes at a replication log. the method also includes asynchronously replicating the plurality of changes from the second table to a first table and maintaining a respective change stream position tracking a respective position in the replication log indicating one or more changes of the plurality of changes that have been replicated. the method includes receiving a read request at the first cluster requesting one or more rows of the first table and returning the respective change stream position.
Inventor(s): Dongeek Shin of San Jose CA (US) for google llc
IPC Code(s): G06F16/958, G06F3/0483
CPC Code(s): G06F16/958
Abstract: according to an aspect, a method may include generating, by an interface-guided diffusion model, a ui-compatible output image in response to one or more prompts including user-generated text and input condition data. the input condition data includes user interface (ui) layout information about at least one ui element included in an interface of an application. the method includes providing the ui-compatible output image to the application.
Inventor(s): Hari Bhaskar Sankaranarayanan of Bangalore (IN) for google llc
IPC Code(s): G06F21/62, G06F21/60
CPC Code(s): G06F21/6245
Abstract: aspects of the disclosed technology include techniques and mechanisms for securing machine learning (ml) training pipelines and ml pipeline models using data loss prevention (dlp). a dlp ml engine may use different levels of detection to determine whether a command manipulates or disseminates sensitive data. the dlp ml engine may progress through the different levels of detection where each level of detection may engage different methods for determining whether the command is used for generative ai purposes. based on determining a command is not used for generative ai purposes, the dlp ml engine may allow execution of the command. alternatively, based on determining the command is used for generative ai purposes, the dlp ml engine may trigger preventive measures.
20240419919. SMART TEXT REWRITING FOR INTERACTIVE DOMAINS_simplified_abstract_(google llc)
Inventor(s): Rongjing Xiang of Sunnyvale CA (US) for google llc, Yue Liang of Sunnyvale CA (US) for google llc, Yang Lu of Mountain View CA (US) for google llc, Xu Liu of Mountain View CA (US) for google llc, Keqian Yan of Shanghai (CN) for google llc, Jiawei Chen of Mountain View CA (US) for google llc
IPC Code(s): G06F40/40, G06F40/166, G06F40/35, G06N3/08, G06T11/60
CPC Code(s): G06F40/40
Abstract: the technology provides rewriting suggestions for communication styles in different pre-defined styles. rewriting may convert text in different ways, including visually enhancing the message. a method provides input to a trained large language model, the input including curated examples associated with one or more writing style choices. the set of curated examples has a first size. the method also includes generating, using the model, a rewriting corpus according to one or more writing style choices. the rewriting corpus has a size two or more orders of magnitude larger than a size of curated examples. the writing style choices include at least one of tone, conversion, application context, or conversation type. a text rewriting model is trained using at least a subset of the rewriting corpus. the model is configured to generate vivid textual information in response to user input in an interactive domain, according to specific writing style choices.
20240419953. QUANTIZED MACHINE-LEARNING CONFIGURATION INFORMATION_simplified_abstract_(google llc)
Inventor(s): Jibing Wang of San Jose CA (US) for google llc, Erik Stauffer of Mountain View CA (US) for google llc
IPC Code(s): G06N3/0495, H04W72/04
CPC Code(s): G06N3/0495
Abstract: aspects describe communicating quantized machine-learning, ml, configuration information over a wireless network. a base station selects () a quantization configuration for quantizing ml configuration information for a deep neural network, dnn, where the quantization configuration indicates one or more quantization formats associated with quantizing the ml configuration information. the base station transmits () an indication of the quantization configuration to a user equipment, ue and transfers (), over the wireless network and with the ue, quantized ml configuration information using the quantization configuration.
Inventor(s): Kevin Satzinger of Goleta CA (US) for google llc, Alexandre Bourassa of Santa Barbara CA (US) for google llc, Dripto Mazumdar Debroy of Los Angeles CA (US) for google llc, Kenneth William Lee of Santa Barbara CA (US) for google llc, Matthew Neeley of Goleta CA (US) for google llc
IPC Code(s): G06N10/40
CPC Code(s): G06N10/40
Abstract: methods, systems and apparatus for dynamically decoupling and performing a target unitary operation to a qubit. in one aspect, a method includes generating a control signal that implements a dynamical decoupling control sequence and applying the control signal to the qubit to dynamically decouple the qubit and perform the target unitary operation on the qubit. the target unitary operation includes a product of multiple sub-unitary operations. the dynamical decoupling control sequence includes a plurality of single qubit gates, where one or more of the single qubit gates comprise a single qubit gate that implements one or more of sub-unitary operations of the multiple sub-unitary operations.
Inventor(s): Yash Parikh of San Carlos CA (US) for google llc, Kwaku Obeng Akoi of Richardson TX (US) for google llc, Jae Min Kim of San Ramon CA (US) for google llc, Neil Mitan Gandhi of Dublin CA (US) for google llc
IPC Code(s): G06Q30/0601, G06K7/14
CPC Code(s): G06Q30/0623
Abstract: the technology is generally directed to providing information associated with a digital component via a visual code after confirming the user scanning the visual code is authorized to access the information. the visual code may be provided for output within or near a digital component on a display device. when the visual code is scanned by an image capturing device, account data associated with the digital component may be compared to profile data of a user profile that is logged into the image capturing device. when the account data corresponds to the profile data, the data associated with the digital component may be provided for display on the image capturing device. when the account data does not correspond to the profile information, a prompt may be provided.
Inventor(s): Clarence Christopher Mysen of San Francisco CA (US) for google llc, Jennifer Liu of Cupertino CA (US) for google llc
IPC Code(s): G06Q30/0601, G06F3/0482, G06F3/0483, G06F16/9535, G06F40/106, G06F40/186, G06Q30/0251
CPC Code(s): G06Q30/0625
Abstract: a content provider can provide content, such as advertisements or other promotional material, to a recipient. a distribution system of the content provider can receive a query and select one or more advertisements and a user interface for presenting the advertisements based on the query, information associated with the user, a set of queries received from the user, and/or other information. a model for selecting the user interface can be generated by serving available user interfaces randomly or pseudo-randomly and monitoring user interaction with the served user interfaces. the model can be updated during regular use based on the performance of the user interfaces.
Inventor(s): Konstantinos Rematas of Zurich (CH) for google llc, Thomas Allen Funkhouser of Menlo CA (US) for google llc, Vittorio Carlo Ferrari of Zurich (CH) for google llc, Andrew Huaming Liu of New York NY (US) for google llc, Andrea Tagliasacchi of Victoria (CA) for google llc, Pratul Preeti Srinivasan of San Francisco CA (US) for google llc, Jonathan Tilton Barron of Alameda CA (US) for google llc
IPC Code(s): G06T15/20, G06N20/00, G06T5/50, G06T5/92, G06T17/10
CPC Code(s): G06T15/205
Abstract: systems and methods for view synthesis and three-dimensional reconstruction can learn an environment by utilizing a plurality of images of an environment and depth data. the use of depth data can be helpful when the quantity of images and different angles may be limited. for example, large outdoor environments can be difficult to learn due to the size, the varying image exposures, and the limited variance in view direction changes. the systems and methods can leverage a plurality of panoramic images and corresponding lidar data to accurately learn a large outdoor environment to then generate view synthesis outputs and three-dimensional reconstruction outputs. training may include the use of an exposure correction network to address lighting exposure differences between training images.
20240420462. Hardware Acceleration of Fingerprint Data Processing_simplified_abstract_(google llc)
Inventor(s): Firas Sammoura of Dublin CA (US) for google llc, James Brooks Miller of Sunnyvale CA (US) for google llc
IPC Code(s): G06V10/94, G06F17/16, G06V10/44, G06V40/12, G06V40/13
CPC Code(s): G06V10/955
Abstract: a method includes identifying, within an image that represents a fingerprint, a plurality of keypoints. the method also includes determining a plurality of pixel vectors by transforming, for each respective keypoint, a corresponding two-dimensional (2d) array of pixel values associated with the respective keypoint into a corresponding pixel vector, and obtaining a plurality of filter vectors that represent, for each respective filter of a plurality of filters, a transformation of a corresponding 2d array of filter values of the respective filter into a corresponding filter vector. the method additionally includes convolving the plurality of keypoints with the plurality of filters by multiplying, using a tensor processing hardware, a pixel matrix formed by the plurality of pixel vectors and a filter matrix formed by the plurality of filter vectors, and determining, based on the convolving, a representation of the fingerprint.
Inventor(s): Te I of San Jose CA (US) for google llc, Chris Kau of Los Altos CA (US) for google llc, Jeffrey Robert Pitman of Santa Clara CA (US) for google llc, Robert Eric Genter of Belmont CA (US) for google llc, Qi Ge of Fremont CA (US) for google llc, Wolfgang Macherey of Sunnyvale CA (US) for google llc, Dirk Ryan Padfield of Seattle WA (US) for google llc, Naveen Arivazhagan of Mountain View CA (US) for google llc, Colin Cherry of Mountain View CA (US) for google llc
IPC Code(s): G10L13/08, G10L13/10, G10L15/00, G10L15/26
CPC Code(s): G10L13/086
Abstract: implementations relate to a multimodal translation application that can provide an abridged version of a translation through an audio interface of a computing device, while simultaneously providing a verbatim textual translation at a display interface of the computing device. the application can provide these different versions of the translation in certain circumstances when, for example, the rate of speech of a person speaking to a user is relatively high compared to a preferred rate of speech of the user. for example, a comparison between phonemes of an original language speech and a translated language speech can be performed to determine whether the ratio satisfies a threshold for providing an audible abridged translation. a determination to provide the abridged translation can additionally or alternatively be based on a determined language of the speaker.
20240420686. SPEECH RECOGNITION WITH SEQUENCE-TO-SEQUENCE MODELS_simplified_abstract_(google llc)
Inventor(s): Rohit Prakash Prabhavalkar of Santa Clara CA (US) for google llc, Zhifeng Chen of Sunnyvale CA (US) for google llc, Bo Li of Fremont CA (US) for google llc, Chung-Cheng Chiu of Sunnyvale CA (US) for google llc, Kanury Kanishka Rao of Santa Clara CA (US) for google llc, Yonghui Wu of Fremont CA (US) for google llc, Ron J. Weiss of New York NY (US) for google llc, Navdeep Jaitly of Mountain View CA (US) for google llc, Michiel A. U. Bacchiani of Summit NJ (US) for google llc, Tara N. Sainath of Jersey City NJ (US) for google llc, Jan Kazimierz Chorowski of POLAND (PL) for google llc, Anjuli Patricia Kannan of Berkeley CA (US) for google llc, Ekaterina Gonina of Sunnyvale CA (US) for google llc, Patrick An Phu Nguyen of Palo Alto CA (US) for google llc
IPC Code(s): G10L15/16, G06N3/08, G10L15/02, G10L15/06, G10L15/22, G10L15/26, G10L25/30
CPC Code(s): G10L15/16
Abstract: a method for performing speech recognition using sequence-to-sequence models includes receiving audio data for an utterance and providing features indicative of acoustic characteristics of the utterance as input to an encoder. the method also includes processing an output of the encoder using an attender to generate a context vector, generating speech recognition scores using the context vector and a decoder trained using a training process, and generating a transcription for the utterance using word elements selected based on the speech recognition scores. the transcription is provided as an output of the asr system.
20240420687. TWO-PASS END TO END SPEECH RECOGNITION_simplified_abstract_(google llc)
Inventor(s): Tara N. Sainath of Jersey City NJ (US) for google llc, Yanzhang He of Palo Alto CA (US) for google llc, Bo Li of Fremont CA (US) for google llc, Arun Narayanan of Milpitas CA (US) for google llc, Ruoming Pang of New York NY (US) for google llc, Antoine Jean Bruguier of Milpitas CA (US) for google llc, Shuo-yiin Chang of Mountain View CA (US) for google llc, Wei Li of Fremont CA (US) for google llc
IPC Code(s): G10L15/16, G06N3/08, G10L15/05, G10L15/06, G10L15/22
CPC Code(s): G10L15/16
Abstract: two-pass automatic speech recognition (asr) models can be used to perform streaming on-device asr to generate a text representation of an utterance captured in audio data. various implementations include a first-pass portion of the asr model used to generate streaming candidate recognition(s) of an utterance captured in audio data. for example, the first-pass portion can include a recurrent neural network transformer (rnn-t) decoder. various implementations include a second-pass portion of the asr model used to revise the streaming candidate recognition(s) of the utterance and generate a text representation of the utterance. for example, the second-pass portion can include a listen attend spell (las) decoder. various implementations include a shared encoder shared between the rnn-t decoder and the las decoder.
Inventor(s): Vikram Aggarwal of Palo Alto CA (US) for google llc, Moises Morgenstern Gali of San Francisco CA (US) for google llc
IPC Code(s): G10L15/18, G10L15/22, G10L15/26, G10L25/78
CPC Code(s): G10L15/1815
Abstract: implementations set forth herein relate providing and/or detecting an automated assistant suppressing output for limiting responsiveness of an automated assistant that is available via another computing device. the output can be rendered as an inaudible tone in an environment, in order to pre-emptively suppress responses from automated assistants that may detect a forthcoming spoken utterance from a user. limiting responsiveness in this way can preserve computational resources in environments, such as vehicles, in which multiple users may be carrying devices that are capable of being invoked via a spoken utterance. for instance, pre-emptively limiting responsiveness of other automated assistants can eliminate interruptions at devices that may other be inadvertently invoked as a result of detecting certain spoken utterances.
Inventor(s): Neeraj Gaur of Mountain View CA (US) for google llc, Tongzhou Chen of Mountain View CA (US) for google llc, Ehsan Variani of Mountain View CA (US) for google llc, Bhuvana Ramabhadran of Mt. Kisco NY (US) for google llc, Parisa Haghani of Mountain View CA (US) for google llc, Pedro J. Moreno Mengibar of Jersey City NJ (US) for google llc
IPC Code(s): G10L15/197, G10L15/00, G10L15/16, G10L15/22
CPC Code(s): G10L15/197
Abstract: a method includes receiving a sequence of acoustic frames extracted from audio data corresponding to an utterance. during a first pass, the method includes processing the sequence of acoustic frames to generate n candidate hypotheses for the utterance. during a second pass, and for each candidate hypothesis, the method includes: generating a respective un-normalized likelihood score; generating a respective external language model score; generating a standalone score that models prior statistics of the corresponding candidate hypothesis; and generating a respective overall score for the candidate hypothesis based on the un-normalized likelihood score, the external language model score, and the standalone score. the method also includes selecting the candidate hypothesis having the highest respective overall score from among the n candidate hypotheses as a final transcription of the utterance.
20240420693. IMPLEMENTATIONS FOR VOICE ASSISTANT ON DEVICES_simplified_abstract_(google llc)
Inventor(s): Kenneth Mixter of Los Altos Hills CA (US) for google llc, Raunaq Shah of Zurich (CH) for google llc
IPC Code(s): G10L15/22, G06F3/16, G10L15/28, G10L15/30, H04M1/27, H04M3/493, H04N21/20, H04N21/239, H04N21/40, H04N21/41, H04N21/4147, H04N21/422, H04N21/45, H04N21/47, H04N21/4722, H04N21/475
CPC Code(s): G10L15/22
Abstract: an electronic device stores a voice assistant library for execution on the electronic device based on the electronic device having a first device type. the electronic device receives a verbal input from a user. it extracts request information from the verbal input by processing the verbal input using the voice assistant library executing on the device. it transmits a request to a remote system. the electronic device receives a response to the request. the response is generated by the remote system. the electronic device performs an operation in accordance with the response by one or more voice-processing modules of the configured voice assistant library.
Inventor(s): Gleb Skobeltsyn of Kilchberg (CH) for google llc, Olga Kapralova of Bern (CH) for google llc, Konstantin Shagin of Adliswil (CH) for google llc, Vladimir Vuskovic of Zollikerberg (CH) for google llc, Yufei Zhao of San Francisco CA (US) for google llc, Bradley Nelson of Mountain View CA (US) for google llc, Alessio Macrì of London (GB) for google llc, Abraham Lee of Belmont CA (US) for google llc
IPC Code(s): G10L15/22, G06F3/16, G10L15/18, G10L15/28
CPC Code(s): G10L15/22
Abstract: implementations described herein relate to providing suggestions, via a display modality, for completing a spoken utterance for an automated assistant, in order to reduce a frequency and/or a length of time that the user will participate in a current and/or subsequent dialog session with the automated assistant. a user request can be compiled from content of an ongoing spoken utterance and content of any selected suggestion elements. when a currently compiled portion of the user request (from content of a selected suggestion(s) and an incomplete spoken utterance) is capable of being performed via the automated assistant, any actions corresponding to the currently compiled portion of the user request can be performed via the automated assistant. furthermore, any further content resulting from performance of the actions, along with any discernible context, can be used for providing further suggestions.
Inventor(s): Denis Burakov of Zurich (CH) for google llc, Sergey Nazarov of Kilchberg (CH) for google llc, Behshad Behzadi of Freienbach (CH) for google llc, Mario Bertschler of Zurich (CH) for google llc, Bohdan Vlasyuk of Zurich (CH) for google llc, Daniel Cotting of Islisberg (CH) for google llc, Michael Golikov of Merlischachen (CH) for google llc, Lucas Mirelmann of Zurich (CH) for google llc, Steve Cheng of Los Altos CA (US) for google llc, Zaheed Sabur of Baar (CH) for google llc, Okan Kolak of Sunnyvale CA (US) for google llc, Yan Zhong of Sunnyvale CA (US) for google llc, Vinh Quoc Ly of Sunnyvale CA (US) for google llc
IPC Code(s): G10L15/22, G06F3/16
CPC Code(s): G10L15/22
Abstract: implementations set forth herein allow a user to access a first application in a foreground of a graphical interface, and simultaneously employ an automated assistant to respond to notifications arising from a second application. the user can provide an input, such as a spoken utterance, while viewing the first application in the foreground in order to respond to notifications from the second application without performing certain intervening steps that can arise under certain circumstances. such intervening steps can include providing a user confirmation, which can be bypassed, and/or time-limited according to a timer, which can be displayed in response to the user providing a responsive input directed at the notification. a period for the timer can be set according to one or more characteristics that are associated with the notification, the user, and/or any other information that can be associated with the user receiving the notification.
20240420698. COMBINING RESPONSES FROM MULTIPLE AUTOMATED ASSISTANTS_simplified_abstract_(google llc)
Inventor(s): Matthew Sharifi of Kilchberg (CH) for google llc, Victor Carbune of Zurich (CH) for google llc
IPC Code(s): G10L15/22, G06F16/245, G06F16/248, G10L15/26, G10L15/30, G10L15/32
CPC Code(s): G10L15/22
Abstract: systems and methods for determining whether to combine responses from multiple automated assistants. an automated assistant may be invoked by a user utterance, followed by a query, which is provided to a plurality of automated assistants. a first response is received from a first automated assistant and a second response is received from a second automated assistant. based on similarity between the responses, a primary automated assistant determines whether to combine the responses into a combined response. once the combined response has been generated, one or more actions are performed in response to the combined response.
Inventor(s): Victor Carbune of Zurich (CH) for google llc, Alvin Abdagic of Zurich (CH) for google llc, Behshad Behzadi of Freienbach (CH) for google llc, Jacopo Sannazzaro Natta of Berkeley CA (US) for google llc, Julia Proskurnia of Zurich (CH) for google llc, Krzysztof Andrzej Goj of Zurich (CH) for google llc, Srikanth Pandiri of Zurich (CH) for google llc, Viesturs Zarins of Zurich (CH) for google llc, Nicolo D'Ercole of Oberrieden (CH) for google llc, Zaheed Sabur of Baar (CH) for google llc, Luv Kothari of Sunnyvale CA (US) for google llc
IPC Code(s): G10L15/26, G06F3/0488, G06N20/00, G10L15/18, G10L15/22
CPC Code(s): G10L15/26
Abstract: systems and methods described herein relate to determining whether to incorporate recognized text, that corresponds to a spoken utterance of a user of a client device, into a transcription displayed at the client device, or to cause an assistant command, that is associated with the transcription and that is based on the recognized text, to be performed by an automated assistant implemented by the client device. the spoken utterance is received during a dictation session between the user and the automated assistant. implementations can process, using automatic speech recognition model(s), audio data that captures the spoken utterance to generate the recognized text. further, implementations can determine whether to incorporate the recognized text into the transcription or cause the assistant command to be performed based on touch input being directed to the transcription, a state of the transcription, and/or audio-based characteristic(s) of the spoken utterance.
20240420700. PROVIDING ANSWERS TO VOICE QUERIES USING USER FEEDBACK_simplified_abstract_(google llc)
Inventor(s): Gabriel Taubman of Brooklyn NY (US) for google llc, Andrew W. Hogue of HO HO Kus NJ (US) for google llc, John J. Lee of Long Island City NY (US) for google llc
IPC Code(s): G10L15/30, G06F16/33, G06F16/9535, G10L15/22, H04M3/493
CPC Code(s): G10L15/30
Abstract: methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a dialog system using user feedback. according to one implementation, a method includes receiving, by a dialog engine and from a computing device, a voice input; determining, by the dialog engine, that the received voice input is classified as feedback to an answer of a question; identifying, by the dialog engine, a predetermined feedback score associated with the feedback; and adjusting a confidence score associated with the question and the answer based on the predetermined feedback score.
Inventor(s): Laurent El Shafey of Mountain View CA (US) for google llc, Hagen Soltau of Yorktown Heights NY (US) for google llc, Izhak Shafran of Portland OR (US) for google llc
IPC Code(s): G10L17/18, G10L15/06, G10L15/22, G10L15/26, G10L15/30
CPC Code(s): G10L17/18
Abstract: methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing audio data using neural networks.
Inventor(s): Xuemei Ouyang of Mountain View CA (US) for google llc, Daniel Jose Fernandes Barros of Mountain View CA (US) for google llc, Dennis Gee-Wai Yee of Mountain View CA (US) for google llc, Ya-ping Kuo of Mountain View CA (US) for google llc, Yungtsung Chen of Mountain View CA (US) for google llc, Po-wei Yeh of Mountain View CA (US) for google llc
IPC Code(s): G10L19/002
CPC Code(s): G10L19/002
Abstract: the present disclosure provides systems and methods for dynamically adjusting a bit-rate of encoded audio data. a buffer may receive audio data encoded at a first bit-rate. an audio data level corresponding to the amount of audio data stored within the buffer may be determined. the audio data level of the buffer may be determined to be within a buffer zone, and a bit-rate adjustment may be initiated after determining the audio data level is within a buffer zone.
Inventor(s): Karl Allan Tore Rudberg of Älvsjö (SE) for google llc, Alessio Bazzica of Järfälla (SE) for google llc
IPC Code(s): G10L21/0208, G06N20/00, G10L21/0364, G10L25/27, G10L25/84
CPC Code(s): G10L21/0208
Abstract: a system includes a memory and a processing device communicably coupled to the memory. the processing device identifies audio data associated with a plurality of input device. the processing devices determines a speech energy level for each input device by providing the audio data as input to a trained model. for each input device, a statistical value associated with the speech energy level is determined. a strongest input device is identified based on the statistical value. in response to determining that the statistical value associated with the speech energy level of the strongest input device satisfies a threshold condition, the processing device updates the gain value of an input device to an estimated target gain value based on the statistical value of the speech energy level of the respective input device.
Inventor(s): Thibault SELLAM of Brooklyn NY (US) for google llc, Ankur BAPNA of New York NY (US) for google llc, Joshua CAMP of London (GB) for google llc, Diana MACKINNON of London (GB) for google llc, Ankur P. PARIKH of New York NY (US) for google llc, Jason RIESA of San Francisco CA (US) for google llc
IPC Code(s): G10L25/69, G10L15/00, G10L15/06, G10L15/16
CPC Code(s): G10L25/69
Abstract: a method for evaluating synthesized speech, including receiving a speech sample in a first language; and determining a rating of the speech sample based on an encoding of the speech sample by an artificial intelligence encoding model, the rating of the speech sample corresponding to a naturalness of the speech sample, wherein the encoding of the speech sample is based on a first training stage of the encoding model using a first set of training data and a second training stage of the encoding model using a second set of training data, the first set of training data includes unlabeled speech audio, unlabeled text, and paired speech audio and text data in the first language and at least one additional language, and the second set of training data includes rated speech audio.
Inventor(s): Frank Stephen Tromp van Diggelen of Carmel-by-the-Sea CA (US) for google llc, Mohammed Khider of Mountain View CA (US) for google llc
IPC Code(s): H01Q3/04, H01Q1/24
CPC Code(s): H01Q3/04
Abstract: example embodiments relate to centimeter-accurate localization using asymmetric antennas. a pair of mobile devices can establish a wireless communication connection with one device acting as a base station and the other being a rover for real-time kinematic (rtk) positioning. when the mobile devices have similar asymmetric antennas for global navigation satellite system (gnss) signal reception, errors caused by the asymmetric antennas can be canceled by combining measurements from both devices. the rover mobile device may display instructions for the user to adjust its orientation to align with the orientation of the base station mobile device. the rover mobile device can then obtain and use measurement or correction data from the base station to determine its position relative to the rover mobile device. such techniques can be performed by complementary model devices to be used to map areas to a high degree of precision (e.g., centimeter level), including elevation.
Inventor(s): Huanyu Chen of Fremont CA (US) for google llc, Brian Huynh of San Jose CA (US) for google llc
IPC Code(s): H01Q21/28, H01Q1/24, H01Q13/18
CPC Code(s): H01Q21/28
Abstract: various arrangements for positioning antennas, grounds, and a decoupling element are described herein. a display is mounted within a first area of an enclosure of the electronic device. a first antenna and second antenna that operate at one or more frequency bands are mounted within the enclosure. a decoupling element is mounted within a cavity area of the enclosure that, at least partially, isolates the first antenna from the second antenna, wherein a position of the decoupling element defines a common edge of a first cavity area that includes the first antenna and a second cavity area that includes the second antenna, and a first length of the decoupling element is less than a second length of the common edge.
Inventor(s): Vadim Sukhomlinov of Santa Clara CA (US) for google llc, Miguel Angel Osorio Lozano of El Dorado Hills CA (US) for google llc, Christopher J. Frantz of Sunnyvale CA (US) for google llc
IPC Code(s): H04L9/32
CPC Code(s): H04L9/3236
Abstract: this document describes techniques and apparatuses directed at accelerating quantum-resistant, cryptographic hash-based signature computations. upon receipt of an input message, one or more processors implements a hash manager. the hash manager is configured to initialize variables, load the input message and initialized variables into an input buffer, and execute a hash-based signature computation. the hash-based signature computation is repeated for a predetermined number of iterations with each iteration involving loading at least a portion of a digest message directly into a configurable position in the input buffer. in so doing, this method of iterative hash computation bypasses memory copies and bus latencies, accelerating quantum-resistant, cryptographic hash-based signature computations.
20240422013. ANONYMOUS EVENT ATTESTATION_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
IPC Code(s): H04L9/32, H04L9/08
CPC Code(s): H04L9/3257
Abstract: methods, systems, and computer media provide attestation tokens that protect the integrity of communications transmitted from client devices, while at the same time avoiding the use of stable device identifiers that could be used to track client devices or their users. in one approach, client devices can receive batches of n device integrity elements from a device integrity computing system, each corresponding to a different public key. the n device elements can be signed by a device integrity computing system. the signing by the device integrity computing system can be signing with a blind signature scheme. client devices can include throttlers imposing limits on the quantity of attestation tokens created by the client device.
20240422309. SELECTION OF PROJECTED MOTION VECTORS_simplified_abstract_(google llc)
Inventor(s): Lin Zheng of Waterloo (CA) for google llc, Yaowu Xu of Saratoga CA (US) for google llc, Lester Lu of Los Angeles CA (US) for google llc, Jingning Han of Santa Clara CA (US) for google llc, Bohan Li of Santa Clara CA (US) for google llc
IPC Code(s): H04N19/105, H04N19/139, H04N19/172, H04N19/573
CPC Code(s): H04N19/105
Abstract: methods, systems and apparatuses are disclosed including computer readable medium storing instructions used to encode or decode a video or a bitstream encodable or decodable using disclosed steps. the steps include reconstructing a first reference frame and a second reference frame for a current frame to be encoded or decoded, projecting motion vectors of the first reference frame and the second reference frame onto pixels of a current reference frame resulting in a first pixel in the current reference frame being associated with a plurality of projected motion vectors, and selecting a first projected motion vector from the plurality of projected motion vectors as a selected motion vector associated with the first pixel to be used for determining a pixel value of the first pixel, the selection based on magnitudes of the respective ones of the plurality of projected motion vectors.
20240422369. HIGHLY EFFICIENT MODEL FOR VIDEO QUALITY ASSESSMENT_simplified_abstract_(google llc)
Inventor(s): Yilin Wang of Sunnyvale CA (US) for google llc, Miao Yin of Somerset NJ (US) for google llc, Qifei Wang of Sunnyvale CA (US) for google llc, Boqing Gong of Bellevue WA (US) for google llc, Neil Aylon Charles Birkbeck of Cruz CA (US) for google llc, Balineedu Chowdary Adsumilli of Sunnyvale CA (US) for google llc
IPC Code(s): H04N21/2343, G06T7/00, H04N19/132, H04N21/466, H04N21/485
CPC Code(s): H04N21/234363
Abstract: a method for generating, for a video stream of a first spatial resolution and a first temporal resolution, a first reduced quality steam of a second spatial resolution and a second reduced-quality stream of a second temporal resolution. a first subset of stps is sampled from the first reduced-quality stream and a second subset of stps is sampled from the second reduced-quality stream. using a machine learning model (mlm) the stps are processed to identify a quality score for each quality-representative stps that are representative of a quality of the video stream. one or more quality-improving actions for the video stream are identified using the quality scores of the quality-representative stps.
20240422435. Microvideo System, Format, and Method of Generation_simplified_abstract_(google llc)
Inventor(s): Wei Hong of Sunnyvale CA (US) for google llc, Radford Ray Juang of Fremont CA (US) for google llc, Marius Renn of Milpitas CA (US) for google llc
IPC Code(s): H04N23/68, H04N23/667, H04N23/69, H04N23/73
CPC Code(s): H04N23/6811
Abstract: an image capture device includes an image sensor configured to capture raw image data, an image signal processor configured to process the raw image data to form a plurality of image frames, and one or more processors to implement a microvideo generation system. the microvideo generation system is configured to perform a trimming operation relating to the plurality of image frames according to one or more image criteria, and generate, based on the trimming operation, a microvideo including a still image and a video in a same data file.
Inventor(s): Chih-Hsiang Wu of Taoyuan City (TW) for google llc
IPC Code(s): H04W24/02, H04W36/00, H04W36/36, H04W60/06, H04W76/15
CPC Code(s): H04W24/02
Abstract: a method in a user device (ue) capable of operating in dual-mode connectivity (dc) with a master node (mn) and a secondary node (sn) of a radio access network (ran) includes receiving, from the ran, a configuration related to a dc procedure and a network-specified condition to be satisfied before the ue applies the configuration (); determining whether a single connectivity (sc) condition of the ue is satisfied (); and, when sc condition is satisfied, inhibiting the ue from applying the configuration ().
Inventor(s): Hui Wang of Buffalo Grove IL (US) for google llc, Jayachandran Chinnakkannu of Santa Clara CA (US) for google llc
IPC Code(s): H04W48/16, H04W12/06, H04W48/18
CPC Code(s): H04W48/16
Abstract: a first user equipment (ue) is configured to establish a tethered connection with a second ue. the first ue obtains authentication information associated with the second ue. the first ue controls, based on the authentication information, access of the second ue to at least one network slice of a plurality of network slices provided by a network. the first ue is also configured to receive a request from the second ue to access a network slice provided by the network. if the network slice is not available at first ue, the first ue sends a request to the network for the network slice. the first ue receives an authentication request associated with the network slice from the network and authenticates the second ue for the network slice.
Inventor(s): Chih-Hsiang Wu of Taoyuan City (TW) for google llc
IPC Code(s): H04W72/30, H04W28/02, H04W76/12
CPC Code(s): H04W72/30
Abstract: to manage transmission of multicast and/or broadcast services (mbs), a base station receives, from a core network (cn), a request to configure a common tunnel associated with an mbs session, via which the base station is to receive mbs data, from the cn, for wireless transmission to multiple user equipment (ues) (). in response to the request, the base station transmits, to the cn, a configuration of the common tunnel ().
Inventor(s): Chih-Hsiang Wu of Taoyuan City (TW) for google llc
IPC Code(s): H04W72/30, H04W76/40
CPC Code(s): H04W72/30
Abstract: a method for managing transmission of multicast and/or broadcast services (mbs) is implemented in a central unit (cu) of a distributed base station that includes the cu and a distributed unit (du). the method includes receiving, from a core network (cn), a request to configure cn-to-bs resources for transmitting downlink (dl) mbs data associated with an mbs session, from the cn for multiple user equipment units (ues) via the distributed base station (); obtaining a configuration for a downlink (dl) tunnel for transmitting the dl mbs data from the cu to the du (); and communicating the dl mbs data between the cn and the du using the cn-to-bs resources and the configuration for the dl tunnel ().
Inventor(s): Chih-Hsiang Wu of Taoyuan City (TW) for google llc
IPC Code(s): H04W72/30, H04W76/40, H04W88/08
CPC Code(s): H04W72/30
Abstract: a method for managing transmission of mbs is implemented in a cu of a distributed base station that includes the cu and a du. the method includes receiving from a core network (cn), a request to configure resources for transmitting downlink (dl) mbs data associated with an mbs session; determining, based at least one of (i) a capability of the du or (ii) a capability of a ue that joined the mbs session, whether the du should transmit the mbs data to the ue over at least one of a radio interface using a ptp or a ptm delivery mechanism; and causing the du to transmit the mbs data in accordance with the determined delivery mechanism.
- GOOGLE LLC
- G02B27/01
- G02B6/42
- CPC G02B27/0172
- Google llc
- G05D1/00
- CPC G05D1/0214
- G06F9/451
- G06F3/044
- G06F3/16
- CPC G06F9/453
- G06F12/0806
- G06F9/50
- G06F9/54
- CPC G06F12/0806
- G06F16/248
- G06F16/2453
- G06F16/2455
- CPC G06F16/248
- G06F16/27
- G06F16/23
- G06F16/25
- G06F16/28
- CPC G06F16/273
- G06F16/958
- G06F3/0483
- CPC G06F16/958
- G06F21/62
- G06F21/60
- CPC G06F21/6245
- G06F40/40
- G06F40/166
- G06F40/35
- G06N3/08
- G06T11/60
- CPC G06F40/40
- G06N3/0495
- H04W72/04
- CPC G06N3/0495
- G06N10/40
- CPC G06N10/40
- G06Q30/0601
- G06K7/14
- CPC G06Q30/0623
- G06F3/0482
- G06F16/9535
- G06F40/106
- G06F40/186
- G06Q30/0251
- CPC G06Q30/0625
- G06T15/20
- G06N20/00
- G06T5/50
- G06T5/92
- G06T17/10
- CPC G06T15/205
- G06V10/94
- G06F17/16
- G06V10/44
- G06V40/12
- G06V40/13
- CPC G06V10/955
- G10L13/08
- G10L13/10
- G10L15/00
- G10L15/26
- CPC G10L13/086
- G10L15/16
- G10L15/02
- G10L15/06
- G10L15/22
- G10L25/30
- CPC G10L15/16
- G10L15/05
- G10L15/18
- G10L25/78
- CPC G10L15/1815
- G10L15/197
- CPC G10L15/197
- G10L15/28
- G10L15/30
- H04M1/27
- H04M3/493
- H04N21/20
- H04N21/239
- H04N21/40
- H04N21/41
- H04N21/4147
- H04N21/422
- H04N21/45
- H04N21/47
- H04N21/4722
- H04N21/475
- CPC G10L15/22
- G06F16/245
- G10L15/32
- G06F3/0488
- CPC G10L15/26
- G06F16/33
- CPC G10L15/30
- G10L17/18
- CPC G10L17/18
- G10L19/002
- CPC G10L19/002
- G10L21/0208
- G10L21/0364
- G10L25/27
- G10L25/84
- CPC G10L21/0208
- G10L25/69
- CPC G10L25/69
- H01Q3/04
- H01Q1/24
- CPC H01Q3/04
- H01Q21/28
- H01Q13/18
- CPC H01Q21/28
- H04L9/32
- CPC H04L9/3236
- H04L9/08
- CPC H04L9/3257
- H04N19/105
- H04N19/139
- H04N19/172
- H04N19/573
- CPC H04N19/105
- H04N21/2343
- G06T7/00
- H04N19/132
- H04N21/466
- H04N21/485
- CPC H04N21/234363
- H04N23/68
- H04N23/667
- H04N23/69
- H04N23/73
- CPC H04N23/6811
- H04W24/02
- H04W36/00
- H04W36/36
- H04W60/06
- H04W76/15
- CPC H04W24/02
- H04W48/16
- H04W12/06
- H04W48/18
- CPC H04W48/16
- H04W72/30
- H04W28/02
- H04W76/12
- CPC H04W72/30
- H04W76/40
- H04W88/08
