Abstract: A system includes first host machines implementing a public-cloud computing environment, wherein at least one of the first host machines includes a resource manager that provides a public-cloud resource interface through which one or more public-cloud clients interact with one or more virtual machines, and second host machines implementing a private-cloud computing environment, wherein at least one of the second host machines includes one or more private-cloud virtual machines, wherein at least one of the first host machines further includes a private-cloud VM resource provider through which the resource manager interacts with the private-cloud virtual machines, wherein the VM resource provider translates requests to perform virtual machine operations from a public-cloud-resource interface to a private-cloud virtual machine interface, and the private-cloud virtual machines perform the requested virtual machine operations in response to receiving the translated requests from the VM resource provider.

Abstract: The document describes systems and techniques directed at three-dimensional, direction-dependent audio for multi-entity telecommunication. In aspects, a remote device receives multi-stream content, including at least one audio stream, from one or more audio-producing entities and obtains orientation information associated with the one or more audio-producing entities. The remote device can then, using the at least one audio stream and the orientation information, provide direction-dependent, three-dimensional audio sufficient to enable a multi-stereo audio output device to reproduce the spatial audio as if the at least one audio stream is originating from a direction, an elevation, and/or a proximity that corresponds to a physical location of the one or more audio-producing entities.

Abstract: This document describes systems and techniques that enable continuous personalization of face authentication. In aspects, an authentication system associated with a network includes an authentication manager. The authentication manager receives an embedding representing image data associated with a user's face. The authentication manager generates a confidence score based on the embedding. Further, the authentication manager updates previously enrolled embeddings with the embedding based on the confidence score, the embedding meeting a clustering confidence threshold. Through such a technique, the authentication manager can alter the previously enrolled embeddings by which a future embedding is used to authenticate the user's face. By so doing, the techniques may provide more-accurate and successful user authentication over time.

Abstract: A method including monitoring, using a standard level of auditing, one or more processes of a VM and, based on monitoring the process(es), detecting aberrant behavior indicating that an attack against the VM is imminent. Based on detecting aberrant behavior indicating that the attack is imminent, the method includes monitoring, using a heightened level of auditing, the process(es), the heightened level of auditing generating log data representative of memory accesses performed by the VM, and notifying a user of the VM that the imminent attack is detected. During the attack against the VM, maintaining the monitoring of the process(es) using the heightened level of auditing, the method includes determining that the attack has concluded and, based on determining that the attack has concluded, processing the log data to determine an action performed by the detected attack; and monitoring, using the standard level of auditing, the process(es).

Abstract: A method includes receiving a content feed that includes audio data corresponding to speech utterances and processing the content feed to generate a semantically-rich, structured document. The structured document includes a transcription of the speech utterances and includes a plurality of words each aligned with a corresponding audio segment of the audio data that indicates a time when the word was recognized in the audio data. During playback of the content feed, the method also includes receiving a query from a user requesting information contained in the content feed and processing, by a large language model, the query and the structured document to generate a response to the query. The response conveys the requested information contained in the content feed. The method also includes providing, for output from a user device associated with the user, the response to the query.

Abstract: Apparatus, systems, methods, and related computer program products for managing demand-response programs and events. The systems disclosed include an energy management system in operation with an intelligent, network-connected thermostat located at a structure. The thermostat controls an HVAC system to cool the structure using a demand response event implementation profile over the demand response event period. The thermostat can also receive a requested change to the setpoint temperatures defined by the demand response event implementation profile and access a determination of an impact on energy shifting that would result if the requested change is incorporated into the demand response event implementation profile. This determination can be communicated to the energy consumer.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for speech synthesis. The methods, systems, and apparatus include actions of obtaining an audio representation of speech of a target speaker, obtaining input text for which speech is to be synthesized in a voice of the target speaker, generating a speaker vector by providing the audio representation to a speaker encoder engine that is trained to distinguish speakers from one another, generating an audio representation of the input text spoken in the voice of the target speaker by providing the input text and the speaker vector to a spectrogram generation engine that is trained using voices of reference speakers to generate audio representations, and providing the audio representation of the input text spoken in the voice of the target speaker for output.

Abstract: A method includes receiving a textual prompt in a first language and obtaining a fine-tuned prompt embedding configured to guide a large language model (LLM) to generate text in a target language from textual prompts in the first language. The method also includes processing, using the LLM, the textual prompt conditioned on the fine-tuned prompt embedding to generate output text in the target language and concatenating the textual prompt and the generated output text to provide an unspoken textual utterance. The method also includes training a multilingual automatic speech recognition (ASR) model to learn how to recognize speech in the target language by injecting the unspoken textual utterance into a text encoder associated with the multilingual ASR model.

Abstract: Techniques and apparatuses are described that enable integrated second factor authentication. These techniques and apparatuses enable the improved security of something you have without the accompanying inconvenience or chance of loss. To do so, a secure physical entity is integrated within a computing device. While this provides the something you have without a need to carry a separate object with you, the something you have also must not be able to be accessed remotely. To prevent remote access physical wires are connected from the secure physical entity to physical structures on the computing device. In this way, a hacker or cyber thief cannot convince an authentication system that the cyber attacker does indeed have the something you have because to do so the attacker must be in physical possession of the computing device.

Abstract: The present document describes a camera module with electrostatic discharge (ESD) protection. In particular, the camera module includes a lightning rod structure, which guides an ESD current to a safe location (e.g., system ground) when a lens retainer of the camera module is hit by an ESD spark. Due to the impact on camera focus tuning and the risk of audio rub and buzz, the lightning rod structure does not physically touch the lens retainer. As such, a gap separates the lightning rod structure from the lens retainer. When the lens retainer is stressed by an ESD spark, the gap is broken down and a conductive path is established to guide the ESD current to the safe location through the lightning rod structure. In this way, the ESD current flows along a controlled path instead of jumping to arbitrary locations, which protects nearby susceptible circuitry.

Abstract: A method includes receiving a set of training utterances each including a non-synthetic speech representation of a corresponding utterance, and for each training utterance, generating a corresponding synthetic speech representation by using a voice conversion model. The non-synthetic speech representation and the synthetic speech representation form a corresponding training utterance pair. At each of a plurality of output steps for each training utterance pair, the method also includes generating, for output by a speech recognition model, a first probability distribution over possible non-synthetic speech recognition hypotheses for the non-synthetic speech representation and a second probability distribution over possible synthetic speech recognition hypotheses for the synthetic speech representation.

Abstract: A method includes receiving a sequence of acoustic frames characterizing one or more utterances as input to a multilingual automated speech recognition (ASR) model. The method also includes generating a higher order feature representation for a corresponding acoustic frame. The method also includes generating a hidden representation based on a sequence of non-blank symbols output by a final softmax layer. The method also includes generating a probability distribution over possible speech recognition hypotheses based on the hidden representation generated by the prediction network at each of the plurality of output steps and the higher order feature representation generated by the encoder at each of the plurality of output steps. The method also includes predicting an end of utterance (EOU) token at an end of each utterance. The method also includes classifying each acoustic frame as either speech, initial silence, intermediate silence, or final silence.

Abstract: Various arrangements for performing wireless device-to-device communication are presented. An audio output device, such as an earbud or pair of earbuds, can establish a connection with an audio source via a first Bluetooth interface that communicates using a Bluetooth communication protocol on a 2.4 GHz Bluetooth frequency band. The audio output device can negotiate that Bluetooth frequency-shifted communication, such as on a 5 or 6 GHz frequency band, is available for use with the audio source. The audio output device may then perform Bluetooth frequency-shifted communication with the audio source such that the audio output device receives an audio stream from the audio source using Bluetooth frequency-shifted communication and the Bluetooth communication protocol.

Abstract: A display system employs multiple micro-electromechanical system (MEMS) mirrors in series to receive collimated light and direct the light to provide light having input angles corresponding to a desired field of view at a point or line at an incoupler (IC) of a waveguide without an optical relay. An initial one or more MEMS mirrors accepts collimated light and generates the scan angles. A last MEMS mirror in the series scans at a range of angles proportional to the scan angles generated by the initial MEMS mirror(s) and directs the scanned light back to a spot or a line at the IC.

Abstract: Implementations described herein relate to receiving user input directed to an automated assistant, processing the user input to determine whether data from a server and/or third-party application is needed to perform certain fulfillment of an assistant command included in the user input, and generating a prompt that requests a user consent to transmitting of a request to the server and/or the third-party application to obtain the data needed to perform the certain fulfillment. In implementations where the user consents, the data can be obtained and utilized to perform the certain fulfillment. In implementations where the user does not consent, client data can be generated locally at a client device and utilized to perform alternate fulfillment of the assistant command. In various implementations, the request transmitted to the server and/or third-party application can be modified based on ambient noise captured when the user input is received.

Abstract: A planar fin for use in a heat sink includes turbulent structures extending from the sides of the planar fin. Each turbulent structure defines a longitudinal axis and having a first edge that is parallel to the longitudinal axis and connected to a planar surface of the fin. Each turbulent structure also includes a second edge opposite the first edged and in free space. The second edge defines a periphery that varies in distance from the first edge along the length of the longitudinal axis. The periphery of each second edge is further shaped such that turbulent flow of a fluid is induced in the flow flowing over the second edge at at least a predefined flow rate.

Abstract: Systems and methods for augmented-reality tutoring can utilize optical character recognition, natural language processing, and/or augmented-reality rendering for providing real-time notifications for completing a determined task. The systems and methods can include utilizing one or more machine-learned models trained for quantitative reasoning and can include providing a plurality of different user interface elements at different times.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for characterizing a gaze position of a user in a query image. One of the methods includes obtaining a query image of a user captured by a camera of a mobile device; obtaining device characteristics data specifying (ii) characteristics of the mobile device, (ii) characteristics of the camera of the mobile device, or (iii) both; and processing a neural network input comprising (i) one or more images derived from the query image and (ii) the device characteristics data using a gaze prediction neural network, wherein the gaze prediction neural network is configured to, at run time and after the gaze prediction neural network has been trained, process the neural network input to generate a neural network output that characterizes a gaze position of the user in the query image.

Abstract: An automated speech recognition (ASR) transcript of at least a portion of a media content is obtained from an ASR tool. Suggested words are received for corrected words of the ASR transcript of the media content. Features are obtained using at least the suggested words or the corrected words. The features include features relating to sound similarities between the suggested words and the corrected words. The features are input into a machine learning (ML) model to obtain a determination regarding a validity of the suggested words. Responsive to the suggested words constituting a valid suggestion, the suggested words are incorporated into the ASR transcript. At least a portion of the ASR transcript is transmitted to a user device in conjunction with at least a portion of the media content.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer-readable media, for speech recognition using multi-dialect and multilingual models. In some implementations, audio data indicating audio characteristics of an utterance is received. Input features determined based on the audio data are provided to a speech recognition model that has been trained to output score indicating the likelihood of linguistic units for each of multiple different language or dialects. The speech recognition model can be one that has been trained using cluster adaptive training. Output that the speech recognition model generated in response to receiving the input features determined based on the audio data is received. A transcription of the utterance generated based on the output of the speech recognition model is provided.