Abstract: Methods, systems, and computer programs are presented for managing audio files of a user to reduce latencies in play start times on local devices. The audio files are stored on cloud storage managed by a server. One method includes processing a plurality of audio files associated with a user, where the processing is configured to create audio snippet files from each of the plurality of audio files. The audio snippet files representing a beginning part of each of the plurality of audio files. The method also includes transmitting the audio snippet files to a client device and detecting a request from the client to begin playing a first audio file from the plurality of audio files of the user. The first audio file being stored on the cloud storage managed by the server.

Abstract: A method for receiving processed information at a remote device is described. The method includes transmitting from the remote device a verbal request to a first information provider and receiving a digital message from the first information provider in response to the transmitted verbal request. The digital message includes a symbolic representation indicator associated with a symbolic representation of the verbal request and data used to control an application. The method also includes transmitting, using the application, the symbolic representation indicator to a second information provider for generating results to be displayed on the remote device.

Abstract: The present invention relates to the field of panoramic still and motion photography. In a first embodiment, a camera apparatus for panoramic photography includes a first image sensor positioned to capture a first image. The first image sensor has a rolling-shutter readout arranged in portrait orientation. The camera apparatus also includes second image sensor positioned to capture a second image. The second image sensor has a rolling-shutter readout arranged in portrait orientation. Finally, the camera apparatus includes a controller configured to signal the second image sensor to start capturing the second image before the first image sensor finishes capturing the first image. At least a portion of the first image is in front of the second image relative to a forward direction of the camera apparatus.

Abstract: The present disclosure provides systems and methods that use machine-learned models, such as deep neural networks, to predict and prevent adverse conditions at structural assets. One example method includes obtaining data descriptive of a plurality of images that depict at least a portion of a geographic area that contains a first structural asset. The plurality of images include at least a first image captured at a first time and a second image captured at a second time that is different than the first time. The method includes inputting data descriptive of at least the first image, the first time, the second image, and the second time into a condition prediction model. The method includes receiving, as an output of the condition prediction model, at least one prediction regarding the occurrence of an adverse condition at the first structural asset during one or more future time periods.

Abstract: Processing stacked data structures is provided. A system receives an input audio signal detected by a sensor of a local computing device, identifies an acoustic signature, and identifies an account corresponding to the signature. The system establishes a session and a profile stack data structure including a first profile layer having policies configured by a third-party device. The system pushes, to the profile stack data structure, a second profile layer retrieved from the account. The system parses the input audio signal to identify a request and a trigger keyword. The system generates, based on the trigger keyword and the second profile layer, a first action data structure compatible with the first profile layer. The system provides the first action data structure for execution. The system disassembles the profile stack data structure to remove the first profile layer or the second profile layer from the profile stack data structure.

Abstract: This document describes techniques using, and devices embodying, wide-field radar-based gesture recognition. These techniques and devices can enable a great breadth of gestures and uses for those gestures, such as gestures to use, control, and interact with computing and non-computing devices, from software applications to refrigerators.

Abstract: A method, apparatus, and electronic device for voice navigation are disclosed. A voice input mechanism 310 may receive a verbal input from a user to a voice user interface program invisible to the user. A processor 104 may identify in a graphical user interface (GUI) a set of GUI items. The processor 104 may convert the set of GUI items to a set of voice searchable indices 400. The processor 104 may correlate a matching GUI item of the set of GUI items to a phonemic representation of the verbal input.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for hotword detection on multiple devices are disclosed. In one aspect, a method includes the actions of receiving, by a computing device, audio data that corresponds to an utterance. The actions further include determining a likelihood that the utterance includes a hotword. The actions further include determining a loudness score for the audio data. The actions further include based on the loudness score, determining an amount of delay time. The actions further include, after the amount of delay time has elapsed, transmitting a signal that indicates that the computing device will initiate speech recognition processing on the audio data.

Abstract: The techniques, methods, systems, and other mechanisms described herein include processes for determining if customized content should be generated, what information to include in the customized content, and when to provide the customized content. In general, a computing system determines that a user intends to travel to a physical venue. The computing system can determine if an entity associated with the physical venue has a web page. The computing system can determine various aspects of a predicted route of travel from the user's present location to the physical venue. The computing system can use location information indicating the user's current location and determine one or more routes of travel to the physical venue. The computing system can compare one or more determined aspects of the predicted route to threshold values to determine if customized content should be generated and presented to the user.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for facilitating and improving push messaging provided to users. In one aspect, a method includes receiving, from a user device, a request to subscribe to a push messaging service, selecting a given phone number that is uniquely associated with the user based on the received request, transmitting, to a user device, a first message that presents visual information at the user device and includes the given phone number, receiving, by the one or more servers, a second message addressed to the given phone number, selecting an additional message that presents additional visual information at the user device based, at least in part, on the message received from the user device and one or more of a geolocation of the mobile device or a time of day, and transmitting the additional message to the user device.

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: This document generally describes systems, methods, devices, and other techniques related to speaker verification, including (i) training a neural network for a speaker verification model, (ii) enrolling users at a client device, and (iii) verifying identities of users based on characteristics of the users' voices. Some implementations include a computer-implemented method. The method can include receiving, at a computing device, data that characterizes an utterance of a user of the computing device. A speaker representation can be generated, at the computing device, for the utterance using a neural network on the computing device. The neural network can be trained based on a plurality of training samples that each: (i) include data that characterizes a first utterance and data that characterizes one or more second utterances, and (ii) are labeled as a matching speakers sample or a non-matching speakers sample.

Abstract: A method includes ingesting event data over a network for a plurality of events obtained by disparate computing resources. Each event is associated with a respective timestamp and one or more ingestion-attributes. The method includes identifying whether the corresponding event is associated with any custom indexing-attributes defined by a user. The method also includes indexing the corresponding event into a data store as structured data based on the respective timestamp, the one or more ingestion-attributes, and any identified custom indexing-attributes. The method includes evicting any of the events of the event data in the data store for a period of time that satisfies an eviction time period threshold. The method also includes retrieving the data from the data store that is associated with the time range, the ingestion-attributes, or the one custom indexing-attributes.

Abstract: A method includes receiving a speech input from a user and obtaining context metadata associated with the speech input. The method also includes generating a raw speech recognition result corresponding to the speech input and selecting a list of one or more denormalizers to apply to the generated raw speech recognition result based on the context metadata associated with the speech input. The generated raw speech recognition result includes normalized text. The method also includes denormalizing the generated raw speech recognition result into denormalized text by applying the list of the one or more denormalizers in sequence to the generated raw speech recognition result.

Abstract: Aspects of the technology employ an integrated location identification and query processing technique. Network scan information is attached to or otherwise combined with a query at a client device and is sent to a web server or other entity for processing. When a query is initiated, the client device determines whether location services are enabled. If not, upon authorization they are enabled and visible network scan data is quickly obtained for transmission with the query. The server may include both a location identification service and a search service, and can split the received information accordingly among these elements in an efficient manner that minimizes data transfers and latency. From the client device's perspective, only one query is transmitted. Relevant results based on the resolved location are received quickly by the client device. This reduces system latency and power consumption, and also minimizes network overhead by eliminating unnecessary packet transmissions.

Abstract: Methods, systems, and media for selecting content to be presented are provided. In some embodiments, the methods comprise: receiving a request for a first video content item from a user device; receiving an indication that the user device is in a background playback mode; determining that the first video content item is not suitable for presentation in the background playback mode based on one or more properties of audio data; based on the determining that the first video content item is not suitable for presentation in the background playback mode, selecting a second video content item that is suitable for presentation in the background playback mode based on the one or more properties of the audio data of the second video content item; and in response to selecting the second video content item, causing the second video content item to be presented by the user device.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating a target sequence from a source sequence. In one aspect, the system includes a recurrent neural network configured to, at each time step, receive am input for the time step and process the input to generate a progress score and a set of output scores; and a subsystem configured to, at each time step, generate the recurrent neural network input and provide the input to the recurrent neural network; determine, from the progress score, whether or not to emit a new output at the time step; and, in response to determining to emit a new output, select an output using the output scores and emit the selected output as the output at a next position in the output order.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for adjusting ordinal positions for content item slots in response to determining that a content item slot with a highest slot ordinal position relative to the other content item slots in a resource is not within a viewport area. Adjusting the adjusting ordinal positions of either the content item slots or content items to be served in the content item slots ensures that a content item slot in the viewport displays a content item with a highest respective ordinal position in a ranking relative to other content items is rendered in the content item slot within the viewport area.

Abstract: Modifying computer program output in a voice or non-text input activated environment is provided. A system can receive audio signals detected by a microphone of a device. The system can parse the audio signal to identify a computer program to invoke. The computer program can identify a dialog data structure. The system can modify the identified dialog data structure to include a content item. The system can provide the modified dialog data structure to a computing device for presentation. The system can validate the dialog data structure output by the computing device for presentation.