Abstract: Systems and methods are disclosed herein for improved per-frequency counting systems that record interactions between individuals and a group of providers while maintaining differential privacy. A protocol may be defined that specifies frequency bins to categorize identifiers corresponding to individuals. A provider may generate a plurality of private sketches, each corresponding to a plurality of frequencies defined in the protocol. Frequency data is determined for each identifier. Identifiers are encoded into the private sketches corresponding to the identifiers' associated frequency. The plurality of private sketches from each provider in the group of providers are combined to generate a deduplicated distribution across the group. In one implementation, the private sketches of each provider are sequentially merged until all sketches have been combined, from which the total distribution can be estimated.

Abstract: An example method includes receiving a data load request requesting loading and partitioning of an unknown quantity of user data for storage at a data storage system. The user data including a partitioning key; a total data size of the user data; a plurality of rows, each row of the plurality of rows associated with a value defined by the partitioning key; and one or more columns. The method also includes identifying one or more storage constraints for the data storage system. The method further includes, after receiving the user data, determining a plurality of partitioning quantiles defining respective ranges of values of the partitioning key based on the user data and the one or more storage constraints for the data storage system; and range partitioning each row of the user data into files based on the value associated with the row defined by the partitioning key, and the respective ranges of the values of the partitioning key defined by the plurality of partitioning quantiles.

Abstract: A method for estimating an embedding capacity includes receiving, at a deterministic reference encoder, a reference audio signal, and determining a reference embedding corresponding to the reference audio signal, the reference embedding having a corresponding embedding dimensionality. The method also includes measuring a first reconstruction loss as a function of the corresponding embedding dimensionality of the reference embedding and obtaining a variational embedding from a variational posterior. The variational embedding has a corresponding embedding dimensionality and a specified capacity. The method also includes measuring a second reconstruction loss as a function of the corresponding embedding dimensionality of the variational embedding and estimating a capacity of the reference embedding by comparing the first measured reconstruction loss for the reference embedding relative to the second measured reconstruction loss for the variational embedding having the specified capacity.

Abstract: The present document describes a passive thermal-control structure for speakers and associated apparatuses and methods. The architecture of the passive thermal-control structure is such that heat is transferred from electronic subsystems of the electronic speaker device to the passive thermal-control structure, which acts as an internal, structural frame of the electronic speaker device and can provide both thermal mitigation and structural stability. The passive thermal-control structure can conduct heat from the electronic subsystems to a housing of the electronic speaker device.

Abstract: A method for context-dependent in-call video codec switching includes receiving an indication that a far-side endpoint includes a capability to switch from an initial video codec to a secondary video codec during a call. During a call with the far-side endpoint, the method includes determining that a performance of an encoder satisfies a codec switching threshold. The codec switching threshold indicates that the performance of the encoder justifies switching from the initial video codec mid-stream of the call. During a call with the far-side endpoint, the method also includes switching from the initial video codec initially negotiated to govern the call to the secondary video codec.

Abstract: This document describes methods and systems for an antenna system integrated with side-keys of an electronic device. The antenna system enables antenna integration in a metal frame using a metal support structure and fastener(s) to route antenna signals around side-key modules embedded in the frame without encountering or causing interference with the side-key modules. By using these techniques to integrate antennas on areas around the side-key modules, more antennas can be implemented on the electronic device, leading to improved capabilities supporting additional wireless standards and a better user experience in terms of improved communication quality.

Abstract: Implementations described herein relate to methods, computing devices, and non-transitory computer-readable media to generate an output image. In some implementations, a method includes estimating depth for an image to obtain a depth. The method further includes generating a focal table for the image that includes parameters that indicate a focal range and at least one of a front slope or a back slope. The method further includes determining if one or more faces are detected in the image. The method further includes, if one or more faces are detected in the image, identifying a respective face bounding box for each face and adjusting the focal table to include the face bounding boxes. The method further includes, if no faces are detected in the image, scaling the focal table. The method further includes, applying blur to the image using the focal table and the depth map to generate an output image.

Abstract: A method includes receiving a transcription of an utterance, processing, using a first model, the transcription to generate a first text segment that represents an initial portion of a response to the utterance, processing, using a TTS system, the first text segment to generate a first synthesized speech representation, and providing, for audible output, the first synthesized speech representation. The method also includes providing, to a second model different from the first model, the transcription and the first text segment, the second model comprising an LLM configured to process the transcription and the first text segment to generate a second text segment that represents a remaining portion of the response to the utterance. The method further includes obtaining a second synthesized speech representation generated from the second text segment, and providing, for audible output by the user device, the second synthesized speech representation.

Abstract: A computer-implemented method includes receiving at a computer server system, from a computing device that is remote from the server system, a string of text that comprises a search query. The method also includes identifying one or more search results that are responsive to the search query, parsing a document that is a target of one of the one or more results, identifying geographical address information from the parsing, generating a specific geographical indicator corresponding to the one search result, and transmitting for use by the computing device, data for automatically generating a navigational application having a destination at the specific geographical indicator.

Abstract: A device may identify a plurality of sources for outputs that the device is configured to provide. The plurality of sources may include at least one of a particular application in the device, an operating system of the device, a particular area within a display of the device, or a particular graphical user interface object. The device may also assign a set of distinct voices to respective sources of the plurality of sources. The device may also receive a request for speech output. The device may also select a particular source that is associated with the requested speech output. The device may also generate speech having particular voice characteristics of a particular voice assigned to the particular source.

Abstract: A method for training a plurality of task neural networks such that the trained task neural networks are compatible with each other is described. The method includes receiving, for each of the plurality of task neural networks, a respective training data set; forming an auxiliary loss function for the plurality of task neural networks, in which the auxiliary loss function ensures that the trained task neural networks would be compatible with each other; and training the plurality of task neural networks to optimize a combined loss function. The combined loss function is a combination of respective task loss functions for the respective machine learning tasks and an auxiliary loss function that encourages compatibility between the task neural networks.

Abstract: Techniques are described for reducing time to decrypt a next encrypted frame in a content stream by optimizing a license/key acquisition process. When requesting content, a key identifier and/or license identifier may be included within a webpage using a link, script, or similar access point. When a client device sends a request for content, the loading of the webpage within the client device includes the embedded key identifier. Access to the key/license identifier at the client device then may initiate a key/license acquisition process by the client device. The key/license may be obtained from a key management device in parallel with, or prior to, downloading of at least a portion of the content stream.

Abstract: This document describes techniques and devices for a radar recognition-aided search. Through use of a radar-based recognition system, gestures made by, and physiological information about, persons can be determined. In the case of physiological information, the techniques can use this information to refine a search. For example, if a person requests a search for a coffee shop, the techniques may refine the search to coffee shops in the direction that the person is walking. In the case of a gesture, the techniques may refine or base a search solely on the gesture. Thus, a search for information about a store, car, or tree can be made responsive to a gesture pointing at the store, car, or tree with or without explicit entry of a search query.

Abstract: Methods, apparatus, and computer readable media related to soliciting feedback from a user regarding one or more content parameters of a suggestion or other content provided by the automated assistant. The user's feedback may be used to influence future suggestions and/or other content subsequently provided, by the automated assistant in future dialog sessions, to the user and/or to other users. In some implementations, content is provided to a user by an automated assistant in a dialog session between the user and the automated assistant—and the automated assistant provides a prompt that solicits user feedback related to the provided content in a future dialog session between the user and the automated assistant. In some of those implementations, the prompt is provided following input from the user and/or output from the automated assistant, in the future dialog session, that is unrelated to the content provided in the previous dialog session.

Abstract: A data center power system includes an electrical power conductor that includes a live conductor surface and is configured to carry electrical power from a power source through a human-occupiable workspace of a data center. The live conductor surface includes a rail configured to mount to a floor of the data center. The data power system further includes a grounded conductor positioned in the human-occupiable workspace apart from the electrical power conductor; a first electrical connector configured to mount to a data center rack that supports a plurality of electronic devices and moveable to electrically contact the live conductor surface of the electrical power conductor; and a second electrical conductor positioned on the rack and configured to electrically contact the grounded conductor.

Abstract: A request from a user to view a video content item may be received, the requesting user being associated with at least one of a set of preferences or a context. A group of similar users may be identified based the set of preferences or the context. A number of candidate video items corresponding to the video content item may be identified. A watch time difference may be determined for each candidate video item of the number of candidate video items. A subset may be determined based on the watch time difference associated with each candidate video items. The candidate video items in the subset may be ranked based on an activity rate associated with the corresponding candidate video item. A candidate video item with the highest ranking may be provided to the user.

Abstract: An automated speech recognition (ASR) model includes a first encoder, a second encoder, and a decoder. The first encoder receives, as input, a sequence of acoustic frames, and generates, at each of a plurality of output steps, a first higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The second encoder receives, as input, the first higher order feature representation generated by the first encoder at each of the plurality of output steps, and generates, at each of the plurality of output steps, a second higher order feature representation for a corresponding first higher order feature frame. The decoder receives, as input, the second higher order feature representation generated by the second encoder at each of the plurality of output steps, and generates, at each of the plurality of time steps, a first probability distribution over possible speech recognition hypotheses.

Abstract: A media items to be shared with users of a content sharing service are identified. Each of the media items corresponds to a video recording generated by a client device that depicts one or more objects corresponding to a real-world event and/or a geographic location. A location of the client device that generated the video recording corresponding to a respective media item of the media items is determined based on image features depicted in a set of frames of the video recording. A request for content associated with at least one of the real-world event and/or the geographic location is received from another client device connected to the content sharing service. The media items and, for each of the media items, an indication of the location of the client device that generated the corresponding video recording are provided in accordance with the request for content.

Abstract: Mechanisms for authenticating a connection between a user device and a streaming media content device comprising: identifying a collection of candidate streaming media content devices based on beacon(s) detected by a user device; receiving an indication that one of the content devices is to be selected for connection to the user device; transmitting instructions to the content devices to broadcast a signal, wherein the signal broadcast by each of the content devices contains a value that indicates an identity of the content device; receiving, from the user device, an audio signal that includes a detected signal; processing the audio signal to extract the value indicating the identity of a particular content device associated with the detected signal contained in the received audio signal; identifying the particular content device based on the determined value; and transmitting an indication that the user device and the particular content device are authenticated to communicate.

Abstract: A foldable device includes a housing and a foldable display device coupled to the housing. The foldable display device includes a foldable display layer, a foldable glass layer on a first side of the foldable display layer, and a sensing device on a second side of the foldable display layer. The sensing device includes a strain gauge layer. The strain gauge layer includes an array of resistors arranged on a piezoresistive substrate. Changes in resistance levels measured by the array of resistors are used to detect faults such as scratches, punctures, cracks and the like in the foldable glass layer. The foldable device may output an alert in response to detection of a fault in the foldable glass layer.