Abstract: Techniques and apparatuses are described that generate radar-based gesture detection events in an ambient compute environment. Compared to other smart devices that rely on a physical user interface, a smart device with a radar system can support ambient computing by providing an eye-free interaction and less cognitively demanding gesture-based user interface. The radar system uses an ambient-computing machine-learned module to quickly recognize gestures performed by a user up to at least two meters away. To improve the false positive rate, a gesture debouncer evaluates class probabilities generated by the ambient-computing machine-learned module. In particular, the gesture debouncer recognizes a gesture if a probability of a gesture class is greater than a first threshold for one or more consecutive frames. The first threshold is determined to balance recall and false positive performance of the radar system.

Abstract: A method of providing emotive text-to-speech includes obtaining input text characterizing a natural language response generated by an assistant LLM to a query input by a user during a conversation between the user and the assistant LLM, and processing, using the assistant LLM, the input text conditioned on an emotion detection task prompt to predict, as output from the assistant LLM, an emotional state of the natural language response. The method also includes determining, based on the emotional state of the natural language response predicted as output from the assistant LLM, an emotional embedding for the input text and instructing a TTS model to process the input text and the emotional embedding to generate a synthesized speech representation of the natural language response conveying the emotional state of the natural language response as specified by the emotional embedding.

Abstract: A media application receives, from a server, an identification of a first composition type from a set of compositions to apply to an initial image captured with a user device. Responsive to one or more people being detected in the initial image, the media application generates a modified image, where the one or more people are removed from the initial image to obtain the modified image. The media application scores at least one candidate position within the modified image based on corresponding composition rules for the first composition type. The media application provides a graphical guide on a viewfinder of the user device to guide a user to capture a final image, wherein the graphical guide indicates a recommended position for the one or more people in the final image.

Abstract: Techniques are described herein for detecting and suppressing commands in media that may trigger another automated assistant. A method includes: determining, for each of a plurality of automated assistant devices in an environment that are each executing at least one automated assistant, an active capability of the automated assistant device; initiating playback of digital media by an automated assistant; in response to initiating playback, processing the digital media to identify an audio segment in the digital media that, upon playback, is expected to trigger activation of at least one automated assistant executing on at least one of the plurality of automated assistant devices in the environment, based on the active capability of the at least one of the plurality of automated assistant devices; and in response to identifying the audio segment in the digital media, modifying the digital media to suppress the activation of the at least one automated assistant.

Abstract: The present disclosure provides a wearable device configured to selectively permit an emission and a reception of light. The wearable device includes a hub board, a sensor, a light blocking structure, and a sensor cover. The hub board includes an interior surface and an exterior surface. Further, the hub board extends longitudinally in an x-direction. The sensor is connected to the exterior surface of the hub board and includes a light emitter extending from the exterior surface of the hub board and a light receiver extending from the exterior surface of the hub board. The light blocking structure is located between the light emitter and the light receiver. Further, the light blocking structure extends transversely from the exterior surface of the hub board in a y-direction. The sensor cover is positioned over the sensor and includes an interior surface and an exterior surface. Further, the sensor cover includes a light allowing portion and a light blocking portion.

Abstract: Content of a first comment corresponding to an electronic document provided by an online document service is received from a first client device associated with a first user. The content of the first comment includes a particular symbol followed by one or more characters that represent a second user. The content of the first comment corresponding to the electronic document is analyzed to determine one or more user addresses. A notification indicating the first comment has been received is sent to a second client device associated with the second user using the determined one or more user addresses.

Abstract: Techniques and apparatuses are described that implement random modulation of charge-pump noise phases to reduce structured noise induced by the charge pump. In an example aspect, a correlated double sampling (CDS) circuit is coupled to a pixel array including at least one pixel circuit. The CDS circuit receives an input signal generated by the at least one pixel circuit from the pixel array and samples a reset component of the input signal during a first sampling time to generate a reset component sample. The first sampling time is at a first offset from a reset control signal and prior to a settling time of the at least one pixel circuit. The CDS circuit samples a signal component of the input signal during a second sampling time to generate a signal component sample and determines an output signal based on the reset component sample and the signal component sample.

Abstract: The present disclosure describes apparatuses and techniques of adaptive PLMN management for varying network conditions. In some aspects, a PLMN manager of a user equipment (UE) searches, as part of a registration procedure, a data repository of PLMN information that includes a PLMN blocked from registration for a duration of time and determines that the PLMN is blocked due to a signal-related condition with a cell of the PLMN (e.g., weak signal). The PLMN manager then determines, during the duration of time, that the signal-related condition between the UE and the cell has improved. In response to the improvement, the PLMN manager unblocks the PLMN before expiration of the duration of time. By so doing, the UE may reattempt registration with the unblocked PLMN at least once before the duration of time expires, which may allow the UE to register more quickly and reestablish network services.

Abstract: A base station can implement a method for scheduling downlink transmissions to a user equipment (UE). The method may be performed by processing hardware and includes transmitting to the UE a control element that includes information related to a first time resource for receiving a first downlink data unit associated with a media access control layer from the base station and a second time resource for receiving a second downlink data unit associated with the media access control layer from the base station (1102). The first time resource and the second time resource are non-consecutive and have different respective timeslot offsets within one or more frames, the information including an index into a table stored at the UE, the table specifying a plurality of candidate timeslot offsets for the first time resource (1104). In addition, the method includes transmitting to the UE the first downlink data unit over the first time resource and the second downlink data unit over the second resource (1106).

Abstract: Systems and methods for generating augmented reality prerenderings can provide the benefit of an augmented reality rendering without requiring the use of user data. Template images can be used instead of user data to protect the user's privacy while enabling the user to see an object or product rendered onto a preferred template image or a variety of template images.

Abstract: A computer-implemented method for automating actions across applications includes providing a user interface of an object within a first application to a user computing system. The method further includes receiving an input from the user computing system via the user interface, where the input is associated with requesting an action to be taken with a second application, and where the action uses content associated with the object of the first application. Additionally, the method includes performing the action with the second application based at least in part on the input being received.

Abstract: Systems and methods of the present disclosure are directed to a method for training a machine-learned semantic matching model. The method can include obtaining a first and second document and a first and second activity log. The method can include determining, based on the first document activity log and the second document activity log, a relation label indicative of whether the documents are related. The method can include inputting the documents into the model to receive a semantic similarity value representing an estimated semantic similarity between the first document and the second document. The method can include evaluating a loss function that evaluates a difference between the relation label and the semantic similarity value. The method can include modifying values of parameters of the model based on the loss function.

Abstract: Features described herein pertain to smart thermostats, and more particularly, an intelligent brightness lock for smart thermostats. A smart thermostat can include a display, an ambient light sensor, and a radar sensor. Using the ambient light sensor, ambient light level values of an environment surrounding the smart thermostat can be measured. Radar data can be received from the radar sensor, and based on the radar data, a determination can be made that a distance between a person and the smart thermostat is less than a predetermined distance. In response, a brightness lock mode is started which causes content to be displayed on the display at a second brightness level that is equal to or greater than a first brightness level.

Abstract: Devices, systems, and methods for transporting or transferring insect pupae in an aqueous solution are described. The system includes a tank containing a solution with insect pupae, a transfer chamber having an outlet at a bottom portion of the chamber, a solution inlet, an evacuation channel, and a depth measurement system. A tube fluidly connects to tank and the transfer chamber. An evacuation device connected to the evacuation channel removes air from the transfer chamber and generates a vacuum therein to draw the solution through the tube into the transfer chamber.

Abstract: A method includes training, using an un-supervised learning technique, an auxiliary ASR model based on a first set of un-transcribed source task speech utterances to determine a first task vector, training, using the un-supervised learning technique, the auxiliary ASR model based on a second set of un-transcribed speech utterances to determine a second task vector, and training, using the un-supervised learning technique, the auxiliary ASR model based on un-transcribed target task speech utterances to determine a target task vector. The method also includes determining a first correlation between the first and target task vectors, determining a second correlation between the second and target task vectors, and adapting parameters of a trained primary ASR model based on the first and second source task vectors and the first and second correlations to teach the primary ASR model to learn how to recognize speech associated with the target task.

Abstract: The technology is generally directed to determining fatigue damages to an electronic component or a hardware system caused by the transport or movement of the electronic component or hardware system. The electronic components may be, for example, data center hardware components. The electronic components are susceptible to fatigue damage resulting from thermal, shock and vibration events. The thermal, shock, and vibration events may occur, for example, during transport or movement of the electronic components. To fully characterize fatigue, the acceleration and stress cycles may be tracked, which correlate with natural phenomena in the global supply chain and data center environment.

Abstract: Ultra-wideband (UWB) tags can be used as part of a high-resolution motion capture system that may not require a cost or a complexity that is typically associated with visually based motion capture systems. The UWB based motion capture uses a bundle of UWB tags, which in a possible implementation, can be affixed to body parts of a user to sense motion of the body parts. The absolute positions of each UWB tag can then be determined by reconstructing a skeletal topology from a Euclidean distance matrix based on inter-tag ranging measurements using handshake signals of a UWB protocol.

Abstract: A method of operating a distributed storage system, the method includes identifying missing chunks of a file. The file is divided into stripes that include data chunks and non-data chunks. The method also includes identifying non-missing chunks available for reconstructing the missing chunks and reconstructing missing data chunks before reconstructing missing non-data chunks using the available non-missing chunks.

Abstract: A wearable device includes a MEMS resonator. The MEMS resonator is configured to generate an output signal that is indicative of a temperature of a portion of the wearable device that contacts a user's skin when the wearable device is worn by the user. The wearable device includes control circuitry communicatively coupled to the MEMS resonator. The control circuitry is configured to demodulate the output signal and a local oscillator signal indicative of a setpoint temperature for the portion of the wearable device. The control circuitry is configured to determine a phase difference between a phase of the demodulated output signal and a phase of the demodulated local oscillator signal. The control circuitry is further configured to determine a temperature of the user's has changed based on the phase difference.

Abstract: Presented are techniques of identifying, processing and displaying data point clusters associated with map information in an efficient manner. Methods and systems are disclosed which process map information to identify clusters of requested data points for display, based on iterative clustering and filtering of the data points. Methods and systems are also disclosed which generate polygons representing the clusters. The amount of data to be processed and/or displayed can be reduced, without loss of any associated information content in a displayed map.

Abstract: Aspects of the present disclosure describe systems and methods for eye-tracking by steering a scan beam in a two-dimensional pattern over a scan region on the eye and detecting light reflected from a plurality of reflection points in the scan region at an angle-sensitive detector. The three-dimensional location of each reflection point is determined by triangulating the instantaneous propagation directions of the scan beam and the reflected signal from that reflection point. Gaze direction for the eye is determined from the locations of the reflection points in three-dimensional space.

Abstract: Aspects of the disclosure are directed to a metadata cache for extending data warehouse features to data lakes. The metadata cache can accelerate query execution by directly accessing unmanaged data from the data lake rather than accessing the data through the data warehouse. The metadata cache can allow for filtering the unmanaged data to improve the speed of retrieving data for executing a query.

Abstract: Aspects of the disclosure are directed to integrating one or more large language models (LLMs) into a cloud database platform, such as a data warehouse. Users of the cloud database platform can provide queries to instruct one or more LLMs to perform generative natural language processing tasks by manipulating or generating text directly in the cloud database platform with a table valued function. Users can provide input to register or generate one or more LLMs of the cloud database platform for performing the natural language processing tasks. Integrating LLMs into the cloud database platform can improve processing capabilities of the LLMs and save computing resources, as specialized LLMs or application-specific API may no longer be necessary.

Abstract: Systems and methods to reduce latency in a graphical environment are described. The systems receives location information of a computing device and identifies content items that satisfy a boundary condition formed from the location information. The system selects content items from categories using a load balancing technique. The system selects, responsive to a request having no keywords, a content item object using values generate with an offline process. The system provides the content item object to the computing device to cause the computing device to render the content item object in the graphical environment.

Abstract: The present disclosure provides systems and methods for synthesizing computer-readable code based on the receipt of input and output examples. A computing system in accordance with the disclosure can be configured to receive a given input and output, access and library of operations, and perform a search of a library of operations (e.g., transpose, slice, norm, etc.) that can be applied to the input. By applying the operations to the input and tracking the results, the computing system may identify an expression comprising one or a combination of operations that when applied to the input generates the output. In this manner, implementations of the disclosure may be used to identify one or more solutions that a user having access to the library of operations may use to generate the output from the input.

Abstract: A method may receive a first three-dimensional model of at least a body part of a person. A method may receive a second three-dimensional model of a wearable device comprising a head-mounted display configured to project a display. A method may predict, by at least one machine-learning model using the first three-dimensional model and the second three-dimensional model as inputs, a plurality of contact points between the first three-dimensional model and the second three-dimensional model. A method may compute a display fit value based on the plurality of contact points.

Abstract: A method includes receiving audio data encoding an utterance and obtaining a set of bias phrases corresponding to a context of the utterance. Each bias phrase includes one or more words. The method also includes processing, using a speech recognition model, acoustic features derived from the audio to generate an output from the speech recognition model. The speech recognition model includes a first encoder configured to receive the acoustic features, a bias encoder configured to receive data indicating the obtained set of bias phrases, a bias encoder, and a decoder configured to determine likelihoods of sequences of speech elements based on output of the first attention module and output of the bias attention module. The method also includes determining a transcript for the utterance based on the likelihoods of sequences of speech elements.

Abstract: Methods, systems, and media for identifying video objects linked to a source video are provided. In some embodiments, the method comprises: identifying demographic attributes corresponding to a first user participating in an online conversation; determining at least one keyword associated with the online conversation, wherein the keyword indicates a topic of the online conversation; identifying a video object based at least on the demographic attributes and the at least one keyword, wherein the video object comprises a portion of a video; causing the identified video object to be presented in a group of video objects on a first user device associated with the first user; receiving an indication that the identified video object has been selected on the first user device for inclusion in a message in the online conversation; and causing the identified video object to be presented on a second user device associated with the second user.

Abstract: Methods and systems for color recommendations for overlaid text in a video are provided herein. A video for upload is received. The video includes a plurality of frames having overlaid content. Data related to the video is provided as input to a machine learning model. One or more outputs obtained from the machine learning model is used to determine a color value that improves one or more presentation characteristics of the overlaid content within the video. A color associated with the determined color value is provided to a user for the overlaid content.

Abstract: A method includes determining a pose of a portion of a user relative to a surface. The surface is associated with a virtual interface generated by a device worn by the user. The method further includes receiving thermal data indicating contact between the portion of the user and a portion of the surface. The method further includes determining a location of the contact relative to the virtual interface using the thermal data. The method further includes determining a type of interaction associated with the contact using the thermal data.

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: Training and/or using both a high-level policy model and a low-level policy model for mobile robot navigation. High-level output generated using the high-level policy model at each iteration indicates a corresponding high-level action for robot movement in navigating to the navigation target. The low-level output generated at each iteration is based on the determined corresponding high-level action for that iteration, and is based on observation(s) for that iteration. The low-level policy model is trained to generate low-level output that defines low-level action(s) that define robot movement more granularly than the high-level action—and to generate low-level action(s) that avoid obstacles and/or that are efficient (e.g., distance and/or time efficiency).

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: In a general aspect, a display panel includes a plurality of pixel groups. A pixel group of the plurality of pixel groups includes a plurality of light emitters of different colors, and a single current source that is multiplexed to sequentially activate light emitters of the plurality of light emitters using a first plurality of selector switches, The display panel further includes a single driver switch for coupling the single current source with the first plurality of selector switches, and a memory for controlling the single driver switch.

Abstract: A wearable computing device includes a body, a first sensor disposed at an upper part of the body, and a second sensor disposed at a lower part of the body. The first sensor detects ambient light in a surrounding environment of the wearable computing device and the second sensor includes one or more light sources and one or more detectors, the one or more light sources emitting light toward a body part of a user when the wearable computing device is worn by the user, and the one or more detectors receiving a reflection of the light emitted toward the body part to generate a signal indicating a biometric of the user. The wearable computing device further includes one or more processors which adjust a current supplied to the one or more light sources based on an amount of ambient light detected by the first sensor.

Abstract: We disclose the generation and training of Generative Adversarial Networks (GAN) to synthesize clinical images with skin conditions. Synthetic images for a pre-specified skin condition are generated, while being able to vary its size, location and the underlying skin color. We demonstrate that the generated images are of high fidelity using objective GAN evaluation metrics. The synthetic images are not only visually similar to real images, but also embody the respective skin conditions. Additionally, synthetic skin images can be used as a data augmentation technique for training a skin condition classifier, and improve the ability of the classifier to detect rare but malignant conditions.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for fine-tuning diffusion-based generative neural networks in compact parameter spaces for text-to-image generation. In one aspect, a method performed by one or more computers for fine-tuning a diffusion-based generative neural network to obtain a fine-tuned version of the diffusion-based generative neural network is described.

Abstract: The present disclosure relates to dynamically scheduling resource requests in a distributed system based on usage quotas. One example method includes identifying usage information for a distributed system including atoms, each atom representing a distinct item used by users of the distributed system; determining that a usage quota associated with the distributed system has been exceeded based on the usage information, the usage quota representing an upper limit for a particular type of usage of the distributed system; receiving a first request for a particular atom requiring invocation of the particular type of usage represented by the usage quota; determining that a second request for a different type of usage of the particular atom is waiting to be processed; and processing the second request for the particular atom before processing the first request.

Abstract: A warp model is determined for a block to be coded by using local motion extension. Motion for an adjacent block is extended locally into the block to determine motion for predicting the block. A motion model used for inter-prediction of an adjacent block that shares at least a portion of a boundary with a current block to be predicted is determined. The motion model is a first translational motion vector or a first warp model. A second translational motion vector for inter-prediction of the current block is determined. A second warp model for the inter-prediction of the current block is determined using the motion model and the second translational motion vector. An inter-prediction process for the current block is performed using the second warp model. For example, the interprediction process may include determining a prediction block for encoding or decoding the current block.

Abstract: In general, the subject matter described in this specification can be embodied in methods, systems, and program products for receiving user input that defines a search query, and providing the search query to a server system. Information that a search engine system determined was responsive to the search query is received at a computing device. The computing device is identified as in a first state, and a first output mode for audibly outputting at least a portion of the information is selected. The first output mode is selected from a collection of the first output mode and a second output mode. The second output mode is selected in response to the computing device being in a second state and is for visually outputting at least the portion of the information and not audibly outputting the at least portion of the information. At least the portion of information is audibly output.

Abstract: A laser display may include a light engine, which includes a laser diode and a temperature sensor configured to measure a temperature of the laser diode. The light engine may be coupled to a controller configured to determine a drive current for the laser diode to generate an optical power based on a threshold of the laser diode and an efficiency of the laser diode. The controller is configured to calculate the threshold and the efficiency of the laser diode at the measured temperature using a threshold model and an efficiency model of the laser diode, which are generated using a calibration process.

Abstract: In general, the subject matter described in this specification can be embodied in methods, systems, and program products for receiving a voice query at a mobile computing device and generating data that represents content of the voice query. The data is provided to a server system. A textual query that has been determined by a speech recognizer at the server system to be a textual form of at least part of the data is received at the mobile computing device. The textual query is determined to include a carrier phrase of one or more words that is reserved by a first third-party application program installed on the computing device. The first third-party application is selected, from a group of one or more third-party applications, to receive all or a part of the textual query. All or a part of the textual query is provided to the selected first application program.

Abstract: A head-worn display (HWD) configured to provide an extended reality (XR) image to a user include includes a flexible circuitry configured to electrically couple a component of the HWD to a processor or power source. Additionally, this flexible circuit includes a heat tab configured to thermally couple the flexible circuit to the component of the HWD. Due to the flexible circuitry being thermally coupled to the component of the HWD, heat generated by the component of the HWD is spread across the surfaces of the flexible circuitry, lowering the temperature of the component.

Abstract: The technology generally relates to disaggregating memory from an application specific integrated circuit (“ASIC”) package. For example, a high-bandwidth memory (“HBM”) optics module package may be connected to an ASIC package via one or more optical links. The HBM optics module package may include HBM dies(s), HBM chiplet(s) and an optical chiplet. The optical chiplet may be configured to connect the HBM optics module to one or more optical fibers that form an optical link with one or more other components of the ASIC package. By including an optical chiplet in the HBM optics module package, the HBM optics module package may be disaggregated from an ASIC package.

Abstract: The present document describes techniques for motion-triggered fast shutter and readout. The techniques described herein reduce motion disparity and rolling shutter effects associated with capturing images of a fast-moving object. Further, these techniques achieve an adaptive system that provides fast shutter speed as needed, minimizes rectifying post processes, and reduces power waste on stationary-like scene areas. In implementations, a camera system includes a pair of synchronized image sensors (e.g., a main sensor and a companion sensor). The companion sensor, running at a higher frame rate and a lower resolution than the main sensor, detects motion and provides motion information (e.g., area of motion and velocity of a moving object) to the main image sensor. The main image sensor uses a dedicated analog-to-digital converter (ADC) circuitry to increase a shutter speed and readout for pixels inside the motion area.

Abstract: A method for providing proactive assistance includes obtaining, by a digital assistant, a contextual event associated with a user of a user device. The method includes determining, using a local large language model (LLM) executing on the user device, a remote LLM prompt confidence. The method includes determining that the remote LLM prompt confidence satisfies a threshold. Based on determining that the remote LLM prompt confidence satisfies the threshold, the method includes generating a remote LLM prompt for a remote LLM executing remote from the user device. The method includes transmitting, to the remote LLM, the remote LLM prompt. The method includes receiving, at the digital assistant, from the remote LLM, response content providing the proactive assistance associated with the contextual event. The method includes providing, for output from the user device, presentation content based on the response content received from the remote LLM.

Abstract: Techniques and apparatuses are described for performing interdependent human behavior detection and/or classification using active acoustic sensing. With active acoustic sensing, multiple human behaviors can be detected and/or classified during a given time period. Interdependent human behavior detection and/or classification involves using the detection and/or classification of a first human behavior to assist with the detection and/or classification of a second human behavior. With interdependent human behavior detection and/or classification, active acoustic sensing can increase the accuracy and/or reliability of human behavior detection and/or classification compared to other single-behavior-based techniques. Furthermore, interdependent human behavior detection and/or classification can be performed using a single type of sensing modality (e.g., active acoustic sensing) and using a single sensor in some implementations.