Abstract: A system, computer-readable storage medium storing at least one program, and a computer-implemented method for identifying availability of media items is presented. A search query is received from a client device of a user. Instances of media items that satisfy the search query and that are available on content sources accessible to the client device of the user are identified. Aggregate information for the media items is determined based on the instances of the media items. The aggregate information for the media items is transmitted to the client device.

Abstract: A user can submit a visual query that includes one or more images with user free-form selected visual features of interest. Various processing techniques such as optical character recognition (OCR) techniques can be used to recognize text (e.g., in the image, surrounding image(s), etc.) and/or various object detection techniques (e.g., machine-learned object detection models, etc.) may be used to detect objects and particular visual features of objects (e.g., dress, sleeves, color, pattern, etc.) within or related to the visual query. Content related to the detected text or object(s) in combination with the user free-form selected visual feature of interest can be identified and potentially provided to a user as search results. As such, aspects of the present disclosure enable the visual search system to more intelligently process a visual query to provide improved search results and content feeds, including search results which are personalized to account for user search intent.

Abstract: A system, apparatus, and method for encoding and decoding a video image having a plurality of frames is disclosed. Encoding and decoding the video image can include selecting, for a current block, a prediction mode from a plurality of prediction modes; identifying, for the current block, a quantization value; selecting, for the current block, a probability distribution from a plurality of probability distributions based on the identified quantization value using a processor; and entropy encoding the selected prediction mode using the selected probability distribution.

Abstract: Techniques are disclosed for training and/or utilizing an alignments and language model (“ALM”) in automatically determining an ALM score corresponding with natural language text generated using a natural language generation model. The natural language text generated using the natural language generation model can be based on a set of structured data. Additionally or alternatively, the ALM can include a fluency model portion and a semantics model portion. The fluency model portion can be used in determining the fluency and/or grammar of the text. The semantics model portion be used in evaluating the content of the natural language text with respect to the content of the structured data.

Abstract: Provided are methods, systems, devices, apparatuses, and tangible non-transitory computer readable media for navigation and geocoding. The disclosed technology can perform operations including accessing semantic tags associated with images. Each of the semantic tags can be associated with features depicted by one of the images. Further, each of the features can be associated with a geographic location. Based on the semantic tags, landmarks that include the features that satisfy entropic criteria can be identified. The entropic criteria can measure a localized prominence of each of the one or more features. A landmark for use in navigation at the location can be selected based on context data associated with a location on a path that includes a plurality of locations. Furthermore, at least one navigational instruction that references the landmark can be generated.

Abstract: An input device, discovery service, and associated discovery and device linking system are provided that allow for adhoc linking of input devices and display devices using discovery key codes. Input devices and display devices are assigned to discovery groups that simplify the process of connecting input devices to display devices for end users.

Abstract: A method for coding a current block using an intra-prediction mode includes obtaining a focal point, the focal point having coordinates (a, b) in a coordinate system; and generating, using first peripheral pixels and second peripheral pixels, a prediction block for the current block, where the first peripheral pixels form a first peripheral pixel line constituting an x-axis, and where the second peripheral pixels form a second peripheral pixel line constituting a y-axis. Generating the prediction block includes, for each location of the prediction block at a location (i, j) of the prediction block, determining at least one of an x-intercept or a y-intercept; and determining a prediction pixel value for the each location of the prediction block using the at least one of the x-intercept or the y-intercept.

Abstract: Implementations are directed to using an assistant to initiate automated telephone calls with entities. Some implementations identify an item of interest, identify a group of entities associated with the item, and initiate the calls with the entities. During a given call with a given entity, the assistant can request a status update regarding the item, and determine a temporal delay before initiating another call with the given entity to request a further status update regarding the item based on information received responsive to the request. Other implementations receive a request to perform an action on behalf of a user, identify a group of entities that can perform the action, and initiate a given call with a given entity. During the given call, the assistant can initiate an additional call with an additional entity, and generate notification(s), for the user, based on result(s) of the given call and/or the additional call.

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.

Abstract: Balancing content distribution between a machine learning model and a statistical model provides a baseline assurance in combination with the benefits of a well-trained machine learning model for content selection. In some implementations, a server receiving requests for a content item assigns a first proportion of the received requests to a first group and assigns remaining requests to a second group. The server uses a machine learning model to select variations of the requested content item for responding to requests assigned to the first group and uses a statistical model to select content variations for requests assigned to the second group. The server obtains performance information, e.g., acceptance rates for the different variations, and compares performance of the different models used for content selection. Audience share assigned to the machine learning model is increased when it outperforms the statistical model and decreased when it underperforms the statistical model.

Abstract: A super-resolution coding mode is described. An encoded image can be decoded from an encoded bitstream stored on a non-transitory computer-readable storage medium. A flag can indicate whether an image was encoded using the super-resolution mode at a first resolution. Responsive to the flag indicating that the image was encoded using the super-resolution mode, bits indicating an amount of scaling of the image are included. The image is decoded from the encoded bitstream to obtain a reconstructed image at the first resolution, and the reconstructed image is upscaled to a second resolution using the amount of scaling to obtain an upscaled reconstructed image. The second resolution is higher than the first resolution. Loop restoration parameters within the bitstream can used for look restoration filtering of the upscaled reconstructed image to obtain a loop restored image at the second resolution.

Abstract: The present disclosure provides practical communication-efficient and low-error algorithms for aggregation of private data. For example, the proposed algorithms can be implemented in the shuffled DP model. Specific example operations that can be performed using the proposed algorithms include summation (e.g., binary summation, integer summation) and histograms over a moderate number of buckets. The proposed algorithms achieve accuracy that is arbitrarily close to that of central DP algorithms with an expected communication per user essentially matching what is needed without any privacy constraints.

Abstract: Techniques of identifying gestures include detecting and classifying inner-wrist muscle motions at a user's wrist using micron-resolution radar sensors. For example, a user of an AR system may wear a band around their wrist. When the user makes a gesture to manipulate a virtual object in the AR system as seen in a head-mounted display (HMD), muscles and ligaments in the user's wrist make small movements on the order of 1-3 mm. The band contains a small radar device that has a transmitter and a number of receivers (e.g., three) of electromagnetic (EM) radiation on a chip (e.g., a Soli chip. This radiation reflects off the wrist muscles and ligaments and is received by the receivers on the chip in the band. The received reflected signal, or signal samples, are then sent to processing circuitry for classification to identify the wrist movement as a gesture.

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.

Abstract: An individual serving size food dispenser includes a casing defining an interior volume configured for receipt of bulk food. An opening of the casing is positioned proximate a bottom portion of the interior volume. An auger is disposed within the interior volume of the casing. The auger includes a first flighting portion and a second flighting portion. The first and second flighting portions are wound in opposite directions. An auger motor is coupled to the auger such that the auger motor is operable to rotate the auger within the interior volume of the casing. A door assembly is positioned proximate the opening of the casing. The door assembly is configured for selectively opening to dispense bulk food from the opening of the casing.

Abstract: Techniques are described herein for enabling more computationally efficient organization of files within a cloud storage system. A method includes: receiving information identifying a document and a set of folders; for each folder in the set of folders, using a trained model to predict a similarity measure between the folder and the document; for each folder in the set of folders, determining a score for the folder based on the predicted similarity measure for the folder; selecting a candidate folder from the set of folders using the scores of the folders within the set of folders; and providing, on a user interface, a selectable option to associate the document with the candidate folder.

Abstract: A system and method of displaying, on a touchscreen display of a mobile device, a virtual game controller having at least one virtual button configured to control multiple actions of an avatar representing a player within a video game executed by a remote gaming server. The player's interaction with the at least one virtual button is received through the touchscreen display and used to generate a control signal based on the received player interaction, the control signal being interpretable by the video game as an instruction to perform multiple actions in the video game. The control signal is transmitted from the mobile device for receipt by the remote gaming server executing the video game.

Abstract: Systems and methods for generating min-increment counting bloom filters to determine count and frequency of device identifiers and attributes in a networking environment are disclosed. The system can maintain a set of data records including device identifiers and attributes associated with device in a network. The system can generate a vector comprising coordinates corresponding to counter registers. The system can identify hash functions to update a counting bloom filter. The system can hash the data records to extract index values pointing to a set of counter registers. The system can increment the positions in the min-increment counting bloom filter corresponding to the minimum values of the counter registers. The system can obtain an aggregated public key comprising a public key. The system can encrypt the counter registers using the aggregated shared key to generate an encrypted vector. The system can transmit the encrypted vector to a networked worker computing device.

Abstract: A system is described that determines, based on information associated with a user of a computing device, an event for initiating an interaction between the user and an assistant executing at the computing device. The system selects, based on the event and from a plurality of actions performed by the assistant, at least one action associated with the event. The system determines, based on the at least one action, whether to output a notification of the event which includes an indication of the event and a request to perform the at least one action associated with the event. Responsive to determining to output the notification of the event, the system sends, to the assistant, the notification of the event for output during the interaction between the user and the assistant.