Abstract: Implementations relate to using deep reinforcement learning to train a model that can be utilized, at each of a plurality of time steps, to determine a corresponding robotic action for completing a robotic task. Implementations additionally or alternatively relate to utilization of such a model in controlling a robot. The robotic action determined at a given time step utilizing such a model can be based on: current sensor data associated with the robot for the given time step, and free-form natural language input provided by a user. The free-form natural language input can direct the robot to accomplish a particular task, optionally with reference to one or more intermediary steps for accomplishing the particular task. For example, the free-form natural language input can direct the robot to navigate to a particular landmark, with reference to one or more intermediary landmarks to be encountered in navigating to the particular landmark.

Abstract: Systems, methods and computer program products are provided for managing contactless transactions. A first tap is performed when a system is placed within a predetermined proximity to a payment terminal. A first select command including an AID corresponding to a first application is received from the payment terminal. A first response based on the first select command is transmitted to the payment terminal. A data request including information indicating supported data types is received from the payment terminal. A second response based on the data request and including transaction data is transmitted to the payment terminal. The transaction data includes at least a portion of commerce data stored in the at least one memory.

Abstract: A system and method are provided for sizing and fitting a head mounted wearable computing device for a user based on image data of the head of the user, including a known reference device having a known scale. The system and method may include capturing image data including a face of the user to be fitted for the head mounted wearable computing device. The known reference device having the known scale is compared to features detected in the image data to determine a scaling factor. The scaling factor is used to size, or assign measures to facial features detected in the image data. A three-dimensional model of the head of the user may be generated from the captured image data.

Abstract: A non-transitory computer-readable storage medium can include instructions stored thereon that, when executed by at least one processor, are configured to cause a computing device to determine, in response to a change in a refresh rate of a display, an encoded intensity of at least a portion of an image presented by the display, determine that the encoded intensity is within a predetermined range, and based on determining that the encoded intensity is within the predetermined range, adjust an intensity of a signal for the portion of the image.

Abstract: Systems and methods for providing audio data, from an initially invoked automated assistant to a subsequently invoked automated assistant. An initially invoked automated assistant may be invoked by a user utterance, followed by audio data that includes a query. The query is provided to a secondary automated assistant for processing. Subsequently, the user can submit a query that is related to the first query. In response, the initially invoked automated assistant provides the query to the secondary automated assistant in lieu of providing the query to other secondary automated assistants based on similarity between the first query and the subsequent query.

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: Implementations determine log-in information indicating whether a user of a first application is logged into a target application and/or a website of the target application. In response to the log-in information indicating that the user is logged into the target application but not logged into the website of the target application, a first selectable element can be displayed at a user interface of the first application to receive user input that causes the target application to be opened. In response to the log-in information indicating that the user is not logged into the target application but is logged into the website of the target application, a second selectable element can be displayed at the first application to receive user input that leads to the website of the target application, or the first selectable element can be assigned a deeplink that leads to the website of the target application.

Abstract: A method includes collecting user activity data for a first online media item. The user activity data can be data for a user consuming the first online media item. The method further includes segmenting a second online media item into a plurality of segments, and identifying one or more of the plurality of segments with user activity data satisfying one or more criteria. The identified segments comprise a set of frames of the plurality of frames of the second online media item. The method further includes selecting a frame from the set of frames from the second online media item. The method further includes sending a recommendation of the selected frame as a thumbnail recommendation for the second online media item to a client device.

Abstract: Implementations are described herein for leveraging digital media files retrieved and/or created by users to predict/determine topics of potential relevance to the users. In various implementations, digital media file(s) created and/or retrieved by a user with a client device may be applied as input across trained machine learning model(s), which in some cases are local to the client device, to generate output that indicates object(s) detected in the digital media file(s). Data indicative of the indicated object(s) may be provided to a remote computing system without providing the digital media file(s) themselves. In some implementations, information associated with the indicated object(s) may be retrieved and proactively output to the user. In some implementations, a frequency at which objects occur across a corpus of digital media files may be considered when determining a likelihood that a detected object is potentially relevant to a user.

Abstract: A method for proactive notifications in a voice interface device includes: receiving a first user voice request for an action with an future performance time; assigning the first user voice request to a voice assistant service for performance; subsequent to the receiving, receiving a second user voice request and in response to the second user voice request initiating a conversation with the user; and during the conversation: receiving a notification from the voice assistant service of performance of the action; triggering a first audible announcement to the user to indicate a transition from the conversation and interrupting the conversation; triggering a second audible announcement to the user to indicate performance of the action; and triggering a third audible announcement to the user to indicate a transition back to the conversation and rejoining the conversation.

Abstract: A method for residual adapters for few-shot text-to-speech speaker adaptation includes obtaining a text-to-speech (TTS) model configured to convert text into representations of synthetic speech, the TTS model pre-trained on an initial training data set. The method further includes augmenting the TTS model with a stack of residual adapters. The method includes receiving an adaption training data set including one or more spoken utterances spoken by a target speaker, each spoken utterance in the adaptation training data set paired with corresponding input text associated with a transcription of the spoken utterance. The method also includes adapting, using the adaption training data set, the TTS model augmented with the stack of residual adapters to learn how to synthesize speech in a voice of the target speaker by optimizing the stack of residual adapters while parameters of the TTS model are frozen.

Abstract: A method includes obtaining a multi-utterance training sample that includes audio data characterizing utterances spoken by two or more different speakers and obtaining ground-truth speaker change intervals indicating time intervals in the audio data where speaker changes among the two or more different speakers occur. The method also includes processing the audio data to generate a sequence of predicted speaker change tokens using a sequence transduction model. For each corresponding predicted speaker change token, the method includes labeling the corresponding predicted speaker change token as correct when the predicted speaker change token overlaps with one of the ground-truth speaker change intervals. The method also includes determining a precision metric of the sequence transduction model based on a number of the predicted speaker change tokens labeled as correct and a total number of the predicted speaker change tokens in the sequence of predicted speaker change tokens.

Abstract: The present disclosure describes techniques and apparatuses that are directed to using memory protection data within a computing device. Techniques include allocating regions of a memory for storing application data and protection data. Techniques also include creating a bitmap having bit values corresponding to memory blocks within the allocated regions. The one or more bit values can be indicative of whether application data and/or protection data are present in a memory block. The techniques and apparatuses can enable memory protection, such as memory security (e.g., encryption) and memory safety (e.g., error correction code (ECC) usage), to be efficiently used while permitting discontiguous memory allocations and without substantial operating system modification.

Abstract: The present disclosure relates to a streaming speech-to-speech conversion model, where an encoder runs in real time while a user is speaking, then after the speaking stops, a decoder generates output audio in real time. A streaming-based approach produces an acceptable delay with minimal loss in conversion quality when compared to other non-streaming server-based models. A hybrid model approach for combines look-ahead in the encoder and a non-causal stacker with non-causal self-attention.

Abstract: Techniques described herein describe aspects of signal adjustments in user equipment-coordination set, UECS, joint transmissions. A base station analyzes a first joint transmission from multiple user equipments, UEs, participating in a UECS, where the multiple UEs include a coordinating UE of the UECS and at least one non-coordinating UE participating in the UECS. The base station determines that the first joint transmission fails to meet a performance metric and directs the multiple UEs participating in the UECS to add signal adjustments to a second joint transmission.

Abstract: This application is directed to a method for controlling user experience (UX) operations on an electronic device that executes an application. A touchless UX operation associated with the application has an initiation condition including at least detection of a presence and a gesture in a required proximity range with a required confidence level. The electronic device then determines from a first sensor signal the proximity of the presence with respect to the electronic device. In accordance with a determination that the determined proximity is in the required proximity range, the electronic device determines from a second sensor signal a gesture associated with the proximity of the presence and an associated confidence level of the determination of the gesture. In accordance with a determination that the determined gesture and associated confidence level satisfy the initiation condition, the electronic device initializes the touchless UX operation associated with the application.

Abstract: A method for identifying malicious software includes receiving and executing a software application, identifying a plurality of uniform resource identifiers the software application interacts with during execution of the software application, and generating a vector representation for the software application using a feed-forward neural network configured to receive the plurality of uniform resource identifiers as feature inputs. The method also includes determining similarity scores for a pool of training applications, each similarity score associated with a corresponding training application and indicating a level of similarity between the vector representation for the software application and a respective vector representation for the corresponding training application.

Abstract: A method includes receiving distillation data including a plurality of out-of-domain training utterances. For each particular out-of-domain training utterance of the distillation data, the method includes generating a corresponding augmented out-of-domain training utterance, and generating, using a teacher ASR model trained on training data corresponding to a target domain, a pseudo-label corresponding to the corresponding augmented out-of-domain training utterance. The method also includes distilling a student ASR model from the teacher ASR model by training the student ASR model using the corresponding augmented out-of-domain training utterances paired with the corresponding pseudo-labels generated by the teacher ASR model.

Abstract: A method includes receiving a sequence of acoustic frames as input to a multilingual automated speech recognition (ASR) model configured to recognize speech in a plurality of different supported languages and generating, by an audio encoder of the multilingual ASR, a higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The method also includes generating, by a language identification (LID) predictor of the multilingual ASR, a language prediction representation for a corresponding higher order feature representation. The method also includes generating, by a decoder of the multilingual ASR, a probability distribution over possible speech recognition results based on the corresponding higher order feature representation, a sequence of non-blank symbols, and a corresponding language prediction representation. The decoder includes monolingual output layer having a plurality of output nodes each sharing a plurality of language-specific wordpiece models.

Abstract: A computer-implemented method for enabling users to subscribe to people and other tagged entities is provided herein. Such a method includes maintaining subscription data specifying a plurality of entities subscribed to by a plurality of users, with each of the plurality of entities being a tagged entity associated with a tag. The method further includes identifying a media item associated with one or more tagged entities of the plurality of entities, determining, based on the subscription data, a user of the plurality of users that is subscribed to the tagged entities of the media item, and providing the media item to the user.