Abstract: A media application obtains a video that includes an audio portion. The media application separates the audio portion into a plurality of channels, where each channel corresponds to a particular audio source. An on-screen classifier model obtains an indication of whether the particular audio source for each channel is depicted in the video. An audio-type classifier model determines, an auditory object classification for each channel. The media application determines a respective gain for each channel based on the indication of whether the particular audio source for the channel is depicted in the video and the auditory object classification for the channel. The media application modifies each channel by applying the respective gain. The media application mixes the modified channels with the audio portion to generate a combined audio.

Abstract: This document describes systems and techniques for implementing personalized suggestions for a user interacting with a facility management system based on contextual metadata to assist the user in controlling the facility management system. For example, a system includes a request module configured to receive a request from a user. A metadata module is configured to access and identify metadata related to a content or context of the request. A large language model (LLM) module is configured to receive the request and the metadata and to generate a suggestion relevant to the content or context of the request. A suggestion module is configured to present the suggestion to the user.

Abstract: A hosted storage system receives a storage request that includes a single object and conforms to an API implemented by the hosted storage system. The API is designed to only support a single object in a storage request. The hosted storage system, in response to determining that the single object is an archive file, extracts each of the bundled files from the archive file and stores each of the extracted files in the hosted storage system such that each of the extracted files is separately accessible by the client system over the network.

Abstract: A method includes receiving a conversational training dataset including a plurality of conversational training samples, each training sample associated with a corresponding conversation and including: corresponding audio data characterizing a corresponding current utterance spoken by a user during a current turn in the corresponding conversation; a corresponding context for the corresponding current utterance including a transcript of a previous turn in the corresponding conversation that precedes the current turn; a corresponding ground-truth transcription of the corresponding current utterance; and a CoT annotation representing a corresponding logical relationship between the corresponding current utterance and the previous turn.

Abstract: A method for distributing encryption keys includes receiving a table associated with a particular user, the table including a plurality of data blocks and splitting the table into a plurality of tablets including a corresponding portion of data blocks. The method also includes generating a resource key uniquely associated with the table and for each tablet generating a unique data encryption key for the corresponding tablet to encrypt with the unique data encryption key. The method also includes encrypting each data encryption key with the resource key and distributing control of each encrypted tablet and each corresponding encrypted data encryption key to a plurality of tablet servers, each controlling one or more of the encrypted tablets. The resource key transmits to a remote entity causing the remote entity to encrypt the resource key with a user key associated with the particular user and transmit the encrypted resource key.

Abstract: A method includes receiving, for presentation to a user of a user device, image data representing an image. The method also includes generating, using a textual story generative model, based on the image data, a textual story for the image, and generating, based on the textual story for the image, textual story audio data representing the textual story for the image. The method further includes providing, for audible output from the user device, the textual story audio data.

Abstract: A method includes receiving an audible response to a query, and prior to playing back the audible response, providing, for output from an acoustic speaker, an alignment output audio stream that encodes an audio watermark. The method also includes receiving an alignment input audio stream captured by a microphone array and encoding an acoustic echo of the audio watermark, processing the alignment input audio stream to detect the acoustic echo, and determining a time alignment value between the alignment output audio stream and the alignment input audio stream. The method also includes playing back a response output audio stream that encodes the audible response and receiving an input audio stream. The input audio stream includes acoustic echo corresponding to the audible response played back. The method also includes processing the input audio stream to generate a respective target audio signal that cancels the acoustic echo.

Abstract: The present disclosure provides systems and methods for transmitting data to a pair of truly wireless devices. A first device, which may be considered the primary device, may determine a first data packet was received. The first device may also determine the first data packet was not received by a second device of the pair of truly wireless devices. In response to determining the second device did not receive the first data packet, the first device may request the first data packet be retransmitted.

Abstract: A computer implemented method includes providing a user interface to a user that includes an original image and an option to generate a high-resolution portion of the original image. The method includes receiving a selection of the option to generate the high-resolution portion of the original image and dimensions of a portion of the original image. The method includes providing the portion of the original image as input to a machine-learning model. The method includes generating, with the machine-learning model, the high-resolution image. The method includes updating the user interface to include the high-resolution portion of the original image.

Abstract: A computer-implemented method includes receiving a natural language command from a user that requests a digital assistant to provide a notification to a user device associated with the user upon occurrence of a particular event. The method also includes processing the natural language command using a natural language understanding module to determine one or more event conditions that each indicate the occurrence of the particular event and obtaining event information of the particular event. While the user device is in a notification silencing mode, the method includes determining that at least one of the one or more event conditions is satisfied and providing the notification for output from the user device in response to determining that at least one of the one or more event conditions is satisfied. The notification, when output from the user device, notifies the user of the occurrence of the particular event.

Abstract: A method includes obtaining labeled training data including a plurality of spoken terms spoken during a conversation. For each respective spoken term, the method includes generating a corresponding sequence of intermediate audio encodings from a corresponding sequence of acoustic frames, generating a corresponding sequence of final audio encodings from the corresponding sequence of intermediate audio encodings, generating a corresponding speech recognition result, and generating a respective speaker token representing a predicted identity of a speaker for each corresponding speech recognition result. The method also includes training the joint speech recognition and speaker diarization model jointly based on a first loss derived from the generated speech recognition results and the corresponding transcriptions and a second loss derived from the generated speaker tokens and the corresponding speaker labels.

Abstract: This document describes systems and techniques for presenting intermediate media to a user that has presented a media request. In aspects, a media request for requested media is received from an input device. The media request is provided to a media service to serve the requested media to a requested device. The media request is processed to identify attributes of the media request indicative of a subject matter of the requested media. Based on the identified attributes of the media request, intermediate media is accessed including one or more images related to the identified attributes. The intermediate media is delivered to the requested device. Accordingly, the intermediate media provides content to engage the user while the user waits for the requested media to be delivered.

Abstract: A method for managing cloud infrastructure includes receiving, from a user of a user device, a cloud infrastructure modification request requesting modification to cloud infrastructure. The cloud infrastructure modification request includes abstract configuration data derived from a user interaction with a graphical user interface (GUI) executing on the user device. The method includes translating the abstract configuration data into a configuration command. The configuration command describes a configuration of the cloud infrastructure. The method includes updating a configuration file with the configuration command. The configuration file includes one or more cloud infrastructure specifications for the cloud infrastructure and is controlled by a source control management system. The method includes provisioning, using the updated configuration file, the cloud infrastructure.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for creating and using virtual email addresses for user privacy protection and data security are described. In one aspect, a method includes receiving, from a user device and at an identity server, login credentials for a first email address mapped to a first set of user information values, detecting, by the identity server, a trigger event, in response to detecting the trigger event, creating, by the identity server, a new virtual email address separate from the first email address and mapped to a second set of user information values different from the first set of user information values, detecting, by the identity server, a request for credentials from a requesting entity, and in response to detecting the request, transmitting, by the identity server, the new virtual email address as new login credentials to the requesting entity.

Abstract: A method includes receiving training data that includes unspoken text utterances, un-transcribed non-synthetic speech utterances, and transcribed non-synthetic speech utterances. Each unspoken text utterance is not paired with any corresponding spoken utterance of non-synthetic speech. Each un-transcribed non-synthetic speech utterance is not paired with a corresponding transcription. Each transcribed non-synthetic speech utterance is paired with a corresponding transcription. The method also includes generating a corresponding synthetic speech representation for each unspoken textual utterance of the received training data using a text-to-speech model. The method also includes pre-training an audio encoder on the synthetic speech representations generated for the unspoken textual utterances, the un-transcribed non-synthetic speech utterances, and the transcribed non-synthetic speech utterances to teach the audio encoder to jointly learn shared speech and text representations.

Abstract: A user equipment (UE) manages thermal levels of antenna modules with reference to a temperature threshold. The UE includes multiple antenna modules having a first antenna module and a second antenna module and at least one wireless transceiver coupled to the multiple antenna modules. The UE also includes a processor and memory system implementing an antenna module thermal manager. The manager is configured to obtain a first temperature indication corresponding to the first antenna module of the multiple antenna modules. The manager is also configured to perform a comparison of the first temperature indication to at least one temperature threshold. The manager is further configured to switch, based on the comparison, from using the first antenna module to using the second antenna module for wireless communication with the at least one wireless transceiver.

Abstract: A method for active learning includes obtaining a set of unlabeled training samples and for each unlabeled training sample, perturbing the unlabeled training sample to generate an augmented training sample. The method includes generating, using a machine learning model, a predicted label for both samples and determining an inconsistency value for the unlabeled training sample that represents variance between the predicted labels for the unlabeled and augmented training samples. The method includes sorting the unlabeled training samples based on the inconsistency values and obtaining, for a threshold number of samples selected from the sorted unlabeled training samples, a ground truth label. The method includes selecting a current set of labeled training samples including each selected unlabeled training samples paired with the corresponding ground truth label. The method includes training, using the current set and a proper subset of unlabeled training samples, the machine learning model.

Abstract: A method can include determining, by a correlation device, that a correlation event has occurred based on at least a first device and a second device having corresponding directional profiles, and responsive to occurrence of the correlation event, causing the first device to transfer a state of the first device by sending a state information message to the second device.

Abstract: A computing system and method can be used to implement a version of federated learning (FL) that incorporates adaptivity (e.g., leverages an adaptive learning rate). In particular, the present disclosure provides a general optimization framework in which (1) clients perform multiple epochs of training using a client optimizer to minimize loss on their local data and (2) a server system updates its global model by applying a gradient-based server optimizer to the average of the clients' model updates. This framework can seamlessly incorporate adaptivity by using adaptive optimizers as client and/or server optimizers. Building upon this general framework, the present disclosure also provides example specific adaptive optimization techniques for FL which use per-coordinate methods as server optimizers. By focusing on adaptive server optimization, the use of adaptive learning rates is enabled without increase in client storage or communication costs and compatibility with cross-device FL can be ensured.

Abstract: Systems and methods for utilizing the dead space around the periphery of a chip for sealing a direct liquid cooled module are disclosed. One of the functions of a direct liquid cooled module is to provide cooling liquid to components located on a chip. A groove member for receiving a sealing member may be applied to the top surface of the chip. The groove member may be directly deposited to the top surface or coupled thereto via an adhesive and/or epoxy. The groove member may be in the form of opposing sidewalls or a u-shaped structure each of which form a partial enclosure for receipt of the sealing member. The groove member may be located entirely within the dead space or at least partially within the dead space and partially within a central area in which the chip components are located. The sealing member may be an O-ring or a gasket.

Abstract: Techniques for initiating download of content are provided, including receiving, by one or more processors of the mobile computing device, an indication of a selection, made by a user of the mobile computing device, associated with third-party content provided by a proximate media player device; identifying, by the one or more processors of the mobile computing device, an indication of downloadable content associated with the third-party content displayed by the media player device; and downloading, by the one or more processors of the mobile computing device, in response to receiving the indication that the user selected the icon and without further input from the user, the downloadable content to a memory of the mobile computing device.

Abstract: Processor(s) of a client device can: receive sensor data that captures environmental attributes of an environment of the client device; process the sensor data using a machine learning model to generate a predicted output that dictates whether one or more currently dormant automated assistant functions are activated; making a decision as to whether to trigger the one or more currently dormant automated assistant functions; subsequent to making the decision, determining that the decision was incorrect; and in response to determining that the determination was incorrect, generating a gradient based on comparing the predicted output to ground truth output. In some implementations, the generated gradient is used, by processor(s) of the client device, to update weights of the on-device speech recognition model. In some implementations, the generated gradient is additionally or alternatively transmitted to a remote system for use in remote updating of global weights of a global speech recognition model.

Abstract: A method for speech conversion includes receiving, as input to an encoder of a speech conversion model, an input spectrogram corresponding to an utterance, the encoder including a stack of self-attention blocks. The method further includes generating, as output from the encoder, an encoded spectrogram and receiving, as input to a spectrogram decoder of the speech conversion model, the encoded spectrogram generated as output from the encoder. The method further includes generating, as output from the spectrogram decoder, an output spectrogram corresponding to a synthesized speech representation of the utterance.

Abstract: A training dataset that includes a first dataset and a second dataset is received. The first dataset includes a first subset of first videos corresponding to a first context and respective first ground truth quality scores of the first videos, and the second dataset includes a second subset of second videos corresponding to a second context and respective second ground truth quality scores of the second videos. A machine learning model is trained to predict the respective first ground truth quality scores and the respective second ground truth quality scores. Training the model includes training it to obtain a global quality score for one of the videos; and training it to map the global quality score to context-dependent predicted quality scores. The context-dependent predicted quality scores include a first context-dependent predicted quality score corresponding to the first context and a second context-dependent predicted quality score corresponding to the second context.

Abstract: Systems and methods of the present disclosure are directed to a method for optimizing utilization of graphics processors for machine learning inference tasks. The method includes simultaneously rendering, by a computing system comprising one or more computing devices, a plurality of textures from an input to a machine-learned model. The method includes generating, by the computing system, a plurality of shaders based at least in part on a layout of the plurality of textures, wherein each of the plurality of shaders corresponds to at least one operator of a plurality of operators of the machine-learned model. The method includes processing, by the computing system using a Graphics Processing Unit (GPU), the plurality of textures with the plurality of shaders to obtain a machine-learning output for the machine-learned model.

Abstract: Implementations set forth herein relate to providing a subsidiary application GUI via a client interface. The GUI can be rendered when a user is accessing a first party system via an application that is provided by the first party or a separate entity. The subsidiary application GUI can be rendered in response to the user providing a search query to the first party server-such as a search query that is in furtherance of initializing receiving certain search results. The server can identify, based on the search query, one or more entities that offer primary and/or subsidiary applications, and request subsidiary data for visibly rendering corresponding subsidiary applications for each entity. The subsidiary applications can optionally provide access to application functions that would not otherwise be available at the client without a corresponding application being installed.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for suppressing search results to personally objectionable content. One of the methods includes receiving an identifier of a resource that has image content. A first classifier classifies the image content as including objectionable content or not including objectionable content. A second classifier classifies the image content as including professionally produced content or not including professionally produced content. Whenever the image content is classified as including objectionable content and as not including professionally produced content, the resource is designated as having personally objectionable content.

Abstract: A method for enhancing dialog systems is disclosed herein. The method may include maintaining an online marketplace that may have a plurality of dialog system extension elements. The plurality of dialog system extension elements may include at least one of a dialog system plugin, a dialog system add-on, a dialog system update, and a dialog system upgrade. The method may further include receiving a selection of one of the plurality of dialog system extension elements from an end user. The end user may be associated with a dialog system. The method may continue with associating the one of the plurality of dialog system extension elements with the dialog system of the end user.

Abstract: Implementations relate to determining a rendering type for an application that is executing automatically. Based on user interactions with an application that is associated with specified input from the user while the user is interacting with the application, a confidence metric is generated for each specified input and a rendering type is determined based on the confidence metrics. Subsequently, when the user requests that a sequence of actions be performed, the application will be displayed according to the rendering type.

Abstract: Techniques and apparatuses are described that implement a smart-device-based radar system capable of determining characteristics of objects external to a vehicle, occupants within a vehicle, and objects proximal to an open-air vehicle. In particular, the system enables a smart device to perform many vehicle operations such as collision avoidance, occupant detection, and parking assistance in vehicle and open-air vehicle environments without integrated radar technology. By using a smart device to perform such actions, existing vehicles and open-air vehicles without integrated radar functionality may be able to leverage radar-based vehicle operations.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating an output sequence from an input sequence. One of the methods includes, at each of a plurality of generation time steps: generating a combined sequence for the generation time step that includes the input sequence followed by the output tokens that have already been generated as of the generation time step; processing the combined sequence using a self-attention decoder neural network to generate a time step output that defines a score distribution over a set of possible output tokens; and selecting, using the time step output, an output token from the set of possible output tokens as the next output token in the output sequence.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network to perform a downstream computer vision task. One of the methods includes pre-training an initial neural network that shares layers with the neural network to perform an initial computer vision task and then training the neural network on the downstream computer vision task.

Abstract: This disclosure relates to a method for managing installation of applications, where the method includes receiving, from a first computing device associated with an administrator of an organization, a pin request to attach an application, provided by an application store platform, with a version of the application, initiating, in response to the pin request, transfer of a copy of the version of the application from the application store platform to a data storage device, generating an application identifier that identifies a location of the version of the application stored in the data storage device, and transmitting installation data to a second computing device that is managed by the organization. The installation data includes the application identifier, which is used by the second computing device to install the version of the application from the data storage device.

Abstract: Adaptive filtering is used video stream for bitrate reduction. A first copy of the input video stream is encoded to a reference bitstream. Each of a number of candidate filters is applied to each frame of a second copy of the input video stream to produce a filtered second copy of the input video stream. The filtered second copy is encoded to a candidate bitstream. A cost value for the candidate filter is determined based on distortion value and bitrate differences between the candidate bitstream and the reference bitstream. The candidate bitstream corresponding to the candidate filter with a lowest one of the cost values is selected as the output bitstream, which is then output or stored. Processing the input video stream using the adaptive filter and before the encoding may result in bitrate reduction, thereby improving compression, decompression, and other performance.

Abstract: In one example, a system comprises a hardware processor that is configured to: receive a request to modify a video content item containing a plurality of media elements; present a user interface that includes a video track representation of the video content item and a layered representation of the plurality of media elements, wherein each media element is represented by a media overlay element positioned proximal to the video track representation, wherein at least one media overlay element overlaps at least another media overlap element to provide a compact representation of the plurality of media elements; and in response to receiving a selected time position within the video track representation, update the layered representation within the user interface to present an expanded overlay list that includes media overlay elements corresponding to the subset of the plurality of media elements that occur at the selected time position.

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: The present disclosure provides systems and methods for calibration-free instant motion tracking useful, for example, for rending virtual content in augmented reality settings. In particular, a computing system can iteratively augment image frames that depict a scene to insert virtual content at an anchor region within the scene, including situations in which the anchor region moves relative to the scene. To do so, the computing system can estimate, for each of a number of sequential image frames: a rotation of an image capture system that captures the image frames; and a translation of the anchor region relative to an image capture system, thereby providing sufficient information to determine where and at what orientation to render the virtual content within the image frame.

Abstract: Methods and apparatus related to identifying a video for completing a task and determining a plurality of video segments of the identified video based on one or more attributes of the task. A task and a plurality of how-to videos related to the task may be identified. A how-to video may be selected and a plurality of video segments of the selected how-to video may be determined. One or more video segments may be associated with one or more task attributes that relate to performing the task. The selected video may be provided to a user and segmented, indexed, and/or annotated based on the associated video segments. In some implementations a given object utilized in performing the task may be identified and one or more video segments corresponding to the given object may be identified and/or provided to the user.

Abstract: Techniques are described herein for enabling the use of “dynamic” or “context-specific” hot words for an automated assistant. In various implementations, an automated assistant may be operated at least in part on a computing device. Audio data captured by a microphone may be monitored for default hot word(s). Detection of one or more of the default hot words may trigger transition of the automated assistant from a limited hot word listening state into a speech recognition state. Transition of the computing device into a given state may be detected, and in response, the audio data captured by the microphone may be monitored for context-specific hot word(s), in addition to or instead of the default hot word(s). Detection of the context-specific hot word(s) may trigger the automated assistant to perform a responsive action associated with the given state, without requiring detection of default hot word(s).

Abstract: A method for representing an intended prosody in synthesized speech includes receiving a text utterance having at least one word, and selecting an utterance embedding for the text utterance. Each word in the text utterance has at least one syllable and each syllable has at least one phoneme. The utterance embedding represents an intended prosody. For each syllable, using the selected utterance embedding, the method also includes: predicting a duration of the syllable by decoding a prosodic syllable embedding for the syllable based on attention by an attention mechanism to linguistic features of each phoneme of the syllable and generating a plurality of fixed-length predicted frames based on the predicted duration for the syllable.

Abstract: Aspects of the disclosure provide a hybrid smartwatch that incorporates digital technology with an analog timepiece in a wristwatch form factor. A digital display layer of a non-emissive material is configured to present notices, data, content and other information. An analog display layer includes one or more hands of the timepiece, and overlies the digital display layer. The hands may be controlled by a processor through micro-stepper motors or other actuators. Physical motion of the hands is simultaneously coupled with arrangement of content or other elements on the display layer, which results in numerous types of hybrid visualizations. This includes temporal presentations using hourly, daily, monthly or other time scales. Shape-shifting of the watch hands between 2D and 1D arrangements can linearly focus on certain information. And various content-aware layouts can highlight, bracket, occlude or otherwise emphasize or deemphasize displayed information.

Abstract: This disclosure relates to deep trench capacitors embedded in a package substrate on which an integrated circuit is mounted. In some aspects, a chip package includes an integrated circuit die that has a power distribution circuit for one or more circuits of the integrated circuit. The chip package also includes a substrate different from the integrated circuit and having a first surface on which the integrated circuit die is mounted and a second surface opposite the first surface. The substrate includes one or more cavities formed in at least one of the first surface or the second surface. The chip package also includes one or more deep trench capacitors disposed in at least one of the one or more cavities. Each deep trench capacitor is connected to the power distribution circuit by conductors.

Abstract: Methods and systems for crowd source-based time marking of media items at a platform are provided herein. A media item is provided to first client devices associated with first users of a platform. An indication is received from each of the first client devices of an interesting content segment of the media item as selected by a first user of the platform. At least one content segment of the media item to be associated with a bookmark for a timeline of the media item is determined. The at least one content segment is determined in view of the indication received from each of the first client devices. The media item and the indication of the bookmark is provided to a second client device for presentation to a second user of the platform.

Abstract: Methods, systems, and media for providing dynamic media sessions are provided.

Abstract: A method including receiving (S605) a request for a depth map, generating (S625) a hybrid depth map based on a device depth map (110) and downloaded depth information (105), and responding (S630) to the request for the depth map with the hybrid depth map (415). The device depth map (110) can be depth data captured on a user device (515) using sensors and/or software. The downloaded depth information (105) can be associated with depth data, map data, image data, and/or the like stored on a remote (to the user device) server (505).