Abstract: Methods, systems, devices, and computer-readable storage media for managing a multi-tenant cloud computing platform. Rather than maintain cluster control planes separately for each tenant, a cloud provider control plane implemented on one or more nodes can manage the creation, allocation, execution, and destruction of cluster control planes in an on-demand way. Computing resources can be used more efficiently at least because resources are not under-utilized. Adding additional tenants can be done with zero or small marginal cost per-tenant, because cluster control planes for servicing requests by the tenant are executed in response to a request, but destroyed when a tenant for the plane is idle. Multiple instances of an API server implementing a platform API for communicating between devices of the platform and clusters of computing resources can be generated and modified on a per-tenant basis.

Abstract: Implementations can reduce the time required to obtain responses from an automated assistant through proactive caching, locally at a client device, of proactive assistant cache entries—and through on-device utilization of the proactive assistant cache entries. Different proactive cache entries can be provided to different client devices, and various implementations relate to technique(s) utilized in determining which proactive cache entries to provide to which client devices. In some of those implementations, in determining which proactive cache entries to provide (proactively or in response to a request) to a given client device, a remote system selects, from a superset of candidate proactive cache entries, a subset of the cache entries for providing to the given client device.

Abstract: Methods, systems, and apparatus for parallel optimization of continuously running quantum error correction by closed-loop feedback. In one aspect, a method includes continuously and effectively optimizing qubit performance in-situ whilst an error correction operation on the quantum system is running. The method directly monitors the output from error detection and provides this information as feedback to calibrate the quantum gates associated with the quantum system. In some implementations, the physical qubits are spatially partitioned into one or more independent hardware patterns, where the errors attributable to each hardware pattern are non-overlapping. The one or more different sets of hardware patterns are then temporarily interleaved such that all physical qubits and operations are optimized. The method allows for the optimization of each section of a hardware pattern to be performed individually and in parallel, and can result is O(1) scaling.

Abstract: Methods, apparatus, systems, and computer-readable media are provided for introducing a user to functions of various applications through interactions with an assistant agent. The assistant agent can correspond to an assistant application that can provide a user interface with multiple selectable elements, each of which can correspond to a separate application. When a user selects one of the selectable elements, a function of an application can be demonstrated to the user, in order that the user might become more familiar with functions of the application. In some implementations, a portion of the selectable element can be selected to cause information about the application to be presented to the user. This allows the user to have the option to try out or learn about an application before investing computational resources through downloading and installing the entire application.

Abstract: A three dimensional neural network accelerator that includes a first neural network accelerator tile that includes a first transmission coil, and a second neural network accelerator tile that includes a second transmission coil, wherein the first neural network accelerator tile is adjacent to and aligned vertically with the second neural network accelerator tile, and wherein the first transmission coil is configured to wirelessly communicate with the second transmission coil via inductive coupling.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for implementing a compressed recurrent neural network (RNN). One of the systems includes a compressed RNN, the compressed RNN comprising a plurality of recurrent layers, wherein each of the recurrent layers has a respective recurrent weight matrix and a respective inter-layer weight matrix, and wherein at least one of recurrent layers is compressed such that a respective recurrent weight matrix of the compressed layer is defined by a first compressed weight matrix and a projection matrix and a respective inter-layer weight matrix of the compressed layer is defined by a second compressed weight matrix and the projection matrix.

Abstract: An example method includes programming, during non-emission periods of frames of a plurality of frames based on image data of the frames, pixels of the plurality of pixels of a display of a computing device; causing, during emission periods of the frames, pixels of the plurality of pixels to emit light, wherein an amount of light emitted by the pixels during an emission period of a particular frame is based on the programming for the particular frame; and synchronizing operation of one or more sensors and operation of the plurality of pixels by at least causing the one or more sensors to alternatingly emit, through the display, electromagnetic radiation during emission periods and non-emission periods.

Abstract: A method at an electronic device with a display includes: displaying a user interface having a first region and a second region; receiving, and displaying in the first region of the user interface, a live video stream of a physical environment captured by a remote video camera; displaying, in the second region, a timeline corresponding to a timespan for a first portion of a duration during which the live video stream may have been recorded; in response to receiving a user interaction to move the timespan to a second portion of the duration, transitioning the displayed timeline to a new timeline that corresponds to the timespan for the second portion, and while transitioning, displaying, in the first region, a subset of video frames representing the first and/or second portion of the duration.

Abstract: PWM-frame rate misalignment is mitigated through implementation of a discrete variable refresh rate (VRR) scheme. A target frame rate is limited to a frame rate selected from only those frame rates that facilitate alignment of each frame period to a specified edge of a PWM cycle of a brightness control signal of a display panel. This alignment results in each frame period at the selected frame rate starting at a same point in a corresponding PWM cycle and ending at a same point in a corresponding PWM cycle to help ensure a constant effective duty cycle across each successive frame period, which in turn mitigates perception of flicker that otherwise would arise. Further, the discrete VRR scheme can employ a compensation mode for compensating for the delay in rendering or otherwise obtaining a frame for display so as to maintain a consistent duty cycle in the brightness control signal.

Abstract: A method for combining hotwords in a single utterance receives, at a first assistant-enabled device (AED), audio data corresponding to an utterance directed toward the first AED and a second AED among two or more AEDs where the audio data includes a query specifying an operation to perform. The method also detects, using a hotword detector, a first hotword assigned to the first AED that is different than a second hotword assigned to the second AED In response to detecting the first hotword, the method initiates processing on the audio data to determine that the audio data includes a term preceding the query that at least partially matches the second hotword assigned. Based on the at least partial match, the method executes a collaboration routine to cause the first AED and the second AED to collaborate with one another to fulfill the query.

Abstract: Remote automated assistant component(s) generate client device notification(s) based on a received IoT state change notification that indicates a change in at least one state associated with at least one IoT device. The generated client device notification(s) can each indicate the change in state associated with the at least one IoT device, and can optionally indicate the at least one IoT device. Further, the remote automated assistant component(s) can identify candidate assistant client devices that are associated with the at least one IoT device, and determine whether each of the one or more of the candidate assistant client device(s) should render a corresponding client device notification.

Abstract: The present disclosure is generally related to inserting supplemental audio content into primary audio content via digital assistant applications. A data processing system can maintain an audio recording of a content publisher and a content spot marker to specify a content spot that defines a time at which to insert supplemental audio content. The data processing system can receive an input audio signal from a client device. The data processing system can parse the input audio signal to determine that the input audio signal corresponds to a request and can identify the audio recording of the content publisher. The data processing system can identify, responsive to the determination, a content selection parameter. The data processing system can select an audio content item using the content selection parameter. The data processing system can generate and transmit an action data structure including the audio recording inserted with audio content item.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining watch time loss regions in media content items. In one aspect, features for a video are input into a trained model that is trained to output watch time loss regions. The trained model is trained using labels corresponding to known watch time loss regions in training videos and features of training videos that correspond to the known watch time loss regions. A watch time loss region defines a time window of a video during which a likelihood of a user stopping playback of the video is more than a threshold likelihood. In response to inputting the feature for the first video into the trained model, data regarding watch time loss regions for the video is obtained from the model and provided to an entity involved in providing the video to a user.

Abstract: A computer-implemented method for optimizing the placement of previously selected breaks in a media item is provided herein. Embodiments of the method include steps of identifying a break in a media item, the break being associated with a first break point at a first time during playback of the media item. The method may also include steps of dynamically adjusting the placement of the breaks within the media item based on the performance of the media item.

Abstract: Routing packetized actions in a voice activated data packet based computer network environment is provided. A system can receive audio signals detected by a microphone of a device. The system can parse the audio signal to identify trigger keyword and request, and generate an action data structure. The system can transmit the action data structure to a third party provider device. The system can receive an indication from the third party provider device that a communication session was established with the device.

Abstract: Implementations described herein relate to a computer-implemented method that includes capturing image data using one or more cameras, wherein the image data includes a primary image and associated depth values. The method further includes encoding the image data in an image format. The encoded image data includes the primary image encoded in the image format and image metadata that includes a device element that includes a profile element indicative of an image type and a first camera element, wherein the first camera element includes an image element and a depth map based on the depth values. The method further includes, after the encoding, storing the image data in a file container based on the image format. The method further includes causing the primary image to be displayed.

Abstract: A method includes a first assistant-enabled device (AED) receiving an assignment instruction assigning a group hotword to a selected group of AEDs that includes the first AED and one or more other AEDs. Each AED is configured to wake-up from a low-power state when the group hotword is detected in streaming audio by at least one of the AEDs. The method also includes receiving audio data that corresponds to an utterance spoken by the user and includes a query that specifies an operation to perform. In response to detecting the group hotword in the audio data, the method also includes triggering the first AED to wake-up from the low-power state and executing a collaboration routine to cause the first AED and each other AED in the selected group of AEDs to collaborate with one another to fulfill performance of the operation specified by the query.

Abstract: The disclosure describes techniques for interacting with visitors at a visitor interaction system. A method includes obtaining video data captured by the visitor interaction system, identifying an approach of a person, and analyzing the video data to determine an identity of the person. The method includes automatically and without user intervention determining that the person belongs to a visitor groups of a plurality of visitor groups. Each of the visitor groups corresponds to a response model that includes a plurality of response actions. At least one of the response actions includes an autonomous response action. The method includes identifying a response model corresponding to the visitor groups and initiating an autonomous response action associated with the response model. Initiating the autonomous response action including presenting a message to the person, monitoring a response to the message, and sending a report message to the user via a client device.

Abstract: A method for a privacy-aware transcription includes receiving audio-visual signal including audio data and image data for a speech environment and a privacy request from a participant in the speech environment where the privacy request indicates a privacy condition of the participant. The method further includes segmenting the audio data into a plurality of segments. For each segment, the method includes determining an identity of a speaker of a corresponding segment of the audio data based on the image data and determining whether the identity of the speaker of the corresponding segment includes the participant associated with the privacy condition. When the identity of the speaker of the corresponding segment includes the participant, the method includes applying the privacy condition to the corresponding segment. The method also includes processing the plurality of segments of the audio data to determine a transcript for the audio data.

Abstract: A panel audio loudspeaker having a panel extends in a plane and an actuator is coupled to the panel. The actuator includes a voice coil attached to and extending from the panel along an axis, a magnet assembly suspended from the panel via one or more springs, and a temperature sensor in electrical contact with the coil at three different axial locations. The temperature sensor is configured, during operation of the device, to measure a temperature of the coil based on voltage measurements at the three different axial locations.

Abstract: A plurality of video items comprising a first subset of video items and a second subset of video items are received at a client device. A graphical user interface comprising a scrollable area to present the plurality of video items and a focus area to present one of the plurality of video items is presented. Video items of the second subset of video items are presented in or adjacent to the focus area when the first subset of video items is visible in the scrollable area and prior to the second subset of video items becoming visible in the scrollable area, and a size of the video items from the second subset of video items presented in or adjacent to the focus area is smaller than a size of video items of the first subset of video items presented in the scrollable area.

Abstract: The subject matter of this specification can be implemented in, among other things, a computer-implemented user interface method including displaying on a touchscreen display a representation of a keyboard defining a top edge and a bottom edge, and a content area adjacent to the keyboard. The method further includes receiving a user dragging input having motion directed to the bottom edge of the keyboard. The method further includes removing the keyboard from the touchscreen display and expanding the content area to an area previously occupied by the keyboard.

Abstract: Methods, systems, and apparatus for performing phase operations. In one aspect, a method for performing a same phase operation on a first and second qubit using a third qubit prepared in a phased plus state includes: performing a first NOT operation on the third qubit; computing a controlled adder operation on the first, second and third qubit, comprising encoding the result of the controlled adder operation in a fourth qubit; performing a square of the phase operation on the fourth qubit; uncomputing the controlled adder operation on the first, second and third qubit; performing a CNOT operation between the first qubit and the third qubit, wherein the first qubit acts as the control; performing a CNOT operation between the second qubit and the third qubit, wherein the second qubit acts as the control; and performing a second NOT operation on the third qubit.

Abstract: Methods, systems, and media for generating search results based on contextual information are provided. In some implementations, a method for presenting search results is provided, the method comprising: receiving, using a hardware processor, a query related to media that is currently being presented; identifying a program that is currently being presented; identifying a plurality of keywords associated with the identified program; determining one or more of the plurality of keywords that are contextually relevant to the query; obtaining a plurality of search results based on the query and the one or more contextually relevant keywords; and causing at least one of the plurality of search results to be presented to the user.

Abstract: Systems, methods, and storage media for generating an online analytical processing cube (MAP) are disclosed. Exemplary implementations may: receive a cube definition file; access a data-source; generate a data-source property configuration for the data-source using the cube definition file to; determine each of respective parsed data from the data-source is a key, attribute, or measure; and generate the OLAP cube by combining the cube definition file and the data-source property configuration for the determined parsed data from the data-source.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for processing an input data element to generate a prediction output that characterizes the input data element. In one aspect, a method comprises: determining a respective attention weight between an input data element and each of a plurality of reference data elements; processing each of the reference data elements using the encoder neural network to generate a respective value embedding of each reference data element; determining a combined value embedding of the reference data elements based on (i) the respective value embedding of each reference data element, and (ii) the respective attention weight between the input data element and each reference data element; and processing the combined value embedding of the reference data elements using a prediction neural network to generate the prediction output that characterizes the input data element.

Abstract: Implementations relate to using deep reinforcement learning to train a model that can be utilized, at each of a plurality of time steps, to determine a corresponding robotic action for completing a robotic task. Implementations additionally or alternatively relate to utilization of such a model in controlling a robot. The robotic action determined at a given time step utilizing such a model can be based on: current sensor data associated with the robot for the given time step, and free-form natural language input provided by a user. The free-form natural language input can direct the robot to accomplish a particular task, optionally with reference to one or more intermediary steps for accomplishing the particular task. For example, the free-form natural language input can direct the robot to navigate to a particular landmark, with reference to one or more intermediary landmarks to be encountered in navigating to the particular landmark.

Abstract: Ranking of geographic information includes receiving information about geographic features, identifying documents based on the information, each document being associated with one of the geographic features, and generating a rank score for each document based on one or more properties of the geographic feature associated with the document.

Abstract: A method of simultaneous multiuser collaborative access to a spreadsheet includes transmitting the spreadsheet to a first device of a first user having a first privilege for a first part that includes a first cell, transmitting the spreadsheet to a second device of a second user having a second privilege for a second part, receiving a first edit of the first cell from the first device, and a second edit of the second cell from the second device, and in response to the first cell and the second cell being different, the first privilege including an edit privilege, and the second privilege including the edit privilege, accepting the first edit, accepting the second edit, forwarding the first edit for display with the first cell at the second device, and providing, to the second device, a real-time visual indication identifying the first user as a last user editing the first part.

Abstract: Systems and methods are disclosed for providing a summary of suggested changes made to an electronic document. A suggested change is iteratively selected from a first list of suggested changes. A second list of modified suggested changes is generated by processing each selected suggested change in the first list of suggested changes. The processing comprises marking the selected suggested change for including or excluding the summary. If the selected suggested change is marked for including the summary, the processing further comprises at least one of combining the selected suggested change with a previously processed suggested change to form one of the modified suggested changes for including in the second list, and truncating the selected suggested change to form one of the modified suggested changes for including in the second list. The second list of modified suggested changes is provided as the summary of suggested changes.

Abstract: Gaze data collected from eye gaze tracking performed while training text was read may be used to train at least one layout interpretation model. In this way, the at least one layout interpretation model may be trained to determine current text that includes words arranged according to a layout, process the current text with the at least one layout interpretation model to determine the layout, and output the current text with the words arranged according to the layout.

Abstract: One or more processors of a mobile computing device may receive, from a view provider, graphical user interface (GUI) view data that specifies, for each respective GUI view of a set of GUI views, a respective platform-neutral layout description and a respective one or more condition for the respective GUI view to be a relevant GUI view. The one or more processors may determine a GUI view as the relevant GUI view out of the set of GUI views based at least in part on one or more conditions for the GUI view specified by the GUI view data. The one or more processors may, in response to determining the GUI view as the relevant GUI view, output, based at least in part on a platform-neutral layout description for the GUI view specified by the GUI view data, the GUI view for display at a display device.

Abstract: A device performs a method for using image data to aid voice recognition. The method includes the device capturing image data of a vicinity of the device and adjusting, based on the image data, a set of parameters for voice recognition performed by the device. The set of parameters for the device performing voice recognition include, but are not limited to: a trigger threshold of a trigger for voice recognition; a set of beamforming parameters; a database for voice recognition; and/or an algorithm for voice recognition. The algorithm may include using noise suppression or using acoustic beamforming.

Abstract: Implementations are directed to facilitating user device and/or agent device actions during a communication session. An interactive communications system provides outputs, as outlined below, that are tailored to enhance the functionality of the communication session, reduce the number of dialog “turns” of the communications session and/or the number of user inputs to devices involved in the session, and/or otherwise mitigate consumption of network and/or hardware resources during the communication session. In various implementations, the communication session involves user device(s) of a user, agent device(s) of an agent, and the interactive communications system. The interactive communications system can analyze various communications from the user device(s) and/or agent device(s) during a communication session in which the user (via the user device(s)) directs various communications to the agent, and in which the agent (via the agent device(s)) optionally directs various communications to the user.

Abstract: A method includes receiving audio data encoding an utterance spoken by a native speaker of a first language, and receiving a biasing term list including one or more terms in a second language different than the first language. The method also includes processing, using a speech recognition model, acoustic features derived from the audio data to generate speech recognition scores for both wordpieces and corresponding phoneme sequences in the first language. The method also includes rescoring the speech recognition scores for the phoneme sequences based on the one or more terms in the biasing term list, and executing, using the speech recognition scores for the wordpieces and the rescored speech recognition scores for the phoneme sequences, a decoding graph to generate a transcription for the utterance.

Abstract: Systems and methods for searching using machine-learned model-generated outputs can provide a user with a medium for generating a theoretical dataset that can then be matched to a real world example. The systems and methods can include selecting a plurality of terms, which can be utilized to generate a prompt input that can be processed by a dataset generation model to generate a plurality of model-generated datasets. A selection can then be received that selects a particular model-generated database to utilize to query a database.

Abstract: This application is directed to a speaker device having a rounded enclosure. The rounded enclosure includes two housing elements that are coupled to each other and have a substantially continuous transition. The rounded enclosure has a circular cross section defined by a first radius. A first housing element includes and extends past the circular cross section, and a second housing element has a second maximum radius that is smaller than the first radius. A speaker unit and a circuit board are arranged within the rounded enclosure. The circuit board is electrically coupled to the speaker unit. At least a portion of the first housing element includes perforations configured to enable transmission of sound generated by the speaker unit out of the speaker device. The second housing element has a power connector that is electrically coupled to the circuit board and is configured to receive power from an external power supply.

Abstract: Techniques described herein relate to facilitating end-to-end multilingual communications with automated assistants. In various implementations, speech recognition output may be generated based on voice input in a first language. A first language intent may be identified based on the speech recognition output and fulfilled in order to generate a first natural language output candidate in the first language. At least part of the speech recognition output may be translated to a second language to generate an at least partial translation, which may then be used to identify a second language intent that is fulfilled to generate a second natural language output candidate in the second language. Scores may be determined for the first and second natural language output candidates, and based on the scores, a natural language output may be selected for presentation.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a classification neural network.

Abstract: A computer-implemented method that includes receiving audio data corresponding to an utterance of a voice command captured by a user device. The user device has a plurality of different users. The method includes determining a particular user among the plurality of different users of the user device as a speaker of the utterance based on a comparison between the audio data and corresponding speaker verification data stored on memory hardware for each user of the plurality of different users of the user device. The method further includes, based on determining the particular user among the plurality of different users of the user device as the speaker of the utterance, providing, for output from the user device, a message comprising a speaker identifier associated with the particular user.

Abstract: In an optical network having a terrestrial terminal and an open cable interface (OCI) connecting a submarine cable to a terrestrial cable, the OCI may include a filter positioned on an optical path between the terrestrial cable and the submarine cable and configured to pass first communication signals of a first frequency band, and filter out secondary signals of a second frequency band that does not overlap with the first frequency band. The secondary signals may be looped back to the terrestrial terminal. The terrestrial terminal may detect the looped back secondary signals, and in response, determine the presence of the OCI and that the supervisory signals were rerouted by the OCI.

Abstract: A method for outsourcing exponentiation in a private group includes executing a query instruction to retrieve a query element stored on an untrusted server by selecting a prime factorization of two or more prime numbers of a modulus associated with the query element stored on the server, obtaining a group element configured to generate a respective one of the prime numbers, generating a series of base values using the prime factorization and the group element, and transmitting the series of base values from the client device to the server. The server is configured to determine an exponentiation of the group element with an exponent stored on the server using the series of base values. The method also includes receiving a result from the server based on the exponentiation of the group element with the exponent.

Abstract: Speech processing techniques are disclosed that enable determining a text representation of alphanumeric sequences in captured audio data. Various implementations include determining a contextual biasing finite state transducer (FST) based on contextual information corresponding to the captured audio data. Additional or alternative implementations include modifying probabilities of one or more candidate recognitions of the alphanumeric sequence using the contextual biasing FST, where the FST further comprises a grammar as well as a speller finite state transducer.

Abstract: In general, techniques are described by which to enable a transparency mode in vehicles. A device comprising one or more microphones and one or more processors may be configured to perform the techniques. The microphones may capture audio data representative of a sound scene external to a vehicle. The processors may perform beamforming with respect to the audio data to obtain object audio data representative of an audio object in the sound scene external to the vehicle. The processors may next reproduce, by interfacing with one or more speakers included within the vehicle and based on the object audio data, the audio object in the sound scene external to the vehicle.

Abstract: According to an aspect, an apparatus may include a media streaming device including electronic circuitry configured to receive media content wirelessly from a media content source, and an output cord segment having a first end portion integrally coupled to a structure of the media streaming device, and a second end portion configured to be coupled to a receiving device, where the electronic circuitry is further configured to transmit the received media content through the output cord segment to the receiving device. The apparatus may include a power cord segment having a first end portion configured to be coupled to the media streaming device, and a second end portion configured to be coupled to a power source.

Abstract: A magnetic bistable hinge system includes a hinge bracket, first and second magnets, and a rotatable member. The first magnet may be coupled to the hinge bracket and be configured to rotate about an axis of the hinge bracket from a first bistable position to a second bistable position. When the south pole of the first magnet is at a first angle relative to the south pole of the second magnet, the first magnet and the second magnets are in the first bistable position and the rotatable member is in a closed position relative to the hinge bracket. When the south pole of the first magnet is at a second angle relative to the south pole of the second magnet, the first and second magnets are at the second bistable position, and the rotatable member is rotated to an open position relative to the hinge bracket.

Abstract: A computer-implemented method is described. The method includes a computing system receiving an item of digital content from a user device. The computing system generates one or more labels that indicate attributes of the item of digital content. The computing system also generates one or more conversational replies to the item of digital content based on the one or more labels that indicate attributes of the item of digital content. The method also includes the computing system selecting a conversational reply from among the one or more conversational replies and providing the conversational reply for output to the user device.

Abstract: A speaker verification method includes receiving audio data corresponding to an utterance, processing a first portion of the audio data that characterizes a predetermined hotword to generate a text-dependent evaluation vector, and generating one or more text-dependent confidence scores. When one of the text-dependent confidence scores satisfies a threshold, the operations include identifying a speaker of the utterance as a respective enrolled user associated with the text-dependent confidence score that satisfies the threshold and initiating performance of an action without performing speaker verification. When none of the text-dependent confidence scores satisfy the threshold, the operations include processing a second portion of the audio data that characterizes a query to generate a text-independent evaluation vector, generating one or more text-independent confidence scores, and determining whether the identity of the speaker of the utterance includes any of the enrolled users.

Abstract: Device installation systems, methods, and media for providing ubiquitous connectivity in outdoor environments are provided.

Abstract: The technology described in this document can be embodied in a computer-implemented method that includes receiving, at a processing system, a first signal including an output of a speaker device and an additional audio signal. The method also includes determining, by the processing system, based at least in part on a model trained to identify the output of the speaker device, that the additional audio signal corresponds to an utterance of a user. The method further includes initiating a reduction in an audio output level of the speaker device based on determining that the additional audio signal corresponds to the utterance of the user.