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 first computing device, audio data that corresponds to an utterance. The actions further include determining a first value corresponding to a likelihood that the utterance includes a hotword. The actions further include receiving a second value corresponding to a likelihood that the utterance includes the hotword, the second value being determined by a second computing device. The actions further include comparing the first value and the second value. The actions further include based on comparing the first value to the second value, initiating speech recognition processing on the audio data.

Abstract: Methods, systems, and media for presenting a suggestion to watch videos are provided. In accordance with some embodiments, a method for presenting a suggestion to watch videos is provided, the method comprising: receiving, from a user device, a message that includes a link to a video; determining whether the link has been selected; in response to determining that the link has not been selected, determining a probability value that indicates a likelihood that a user associated with the user device would be interested in viewing the video; in response to determining that the probability value is greater than a predetermined threshold value, determining a time to suggest that the user view the video based at least in part on information relating to movement of the user device; and causing a suggestion to view the video to be presented in a suggestion interface on the user device at the determined time.

Abstract: Methods for ranking users based on contextual factors are provided. An example method may include receiving an online question by an asking user. The method may further include identifying one or more topics of the received online question. The method may further include identifying a set of candidate answering users from a plurality of users based on an expertise level of each user of the plurality of users, wherein the expertise level is related to the identified one or more topics of the received online question. In addition, the method may include ranking the set of candidate answering users based on a commonality of one or more contextual factors between the asking user and the set of candidate answering users, wherein the one or more contextual factors are distinct from the identified one or more topics. Systems and computer-readable media for ranking users based on contextual factors are also described.

Abstract: Methods and apparatus for controlling one or more devices using a computing device, such as a wearable computing device, are disclosed. Using a sensor associated with a wearable computing device, one or more inputs are generated at the wearable computing device. As one example, the sensor can be an image capture device, and the inputs can be one or more images from a point of view of a person associated with the computing device. As another example, the sensor can be a motion sensor, and the input can be locations and/or orientations, and a gaze direction can be determined. The computing device can determine an object of interest based on the inputs. A control signal can be generated for the object of interest. The control signal can be transmitted from the computing device.

Abstract: Systems and methods are described for receiving image content from an emissive display toward a first filter stack, the first filter stack adapted to be oriented in a first direction from an optical axis of a first lens, and toward the first lens, transmitting the image content through a curved lens parallel to the optical axis of the first lens, wherein the curved lens transmits a portion of the image content to at least one optical element and to a second filter stack, the second filter stack being adapted to be oriented in a second direction from the optical axis of the first lens, and receiving the portion from the second filter stack and providing at least some of the portion to the first lens for viewing by a user.

Abstract: Methods, systems, and apparatus for determining resource quality based on resource competition. In an aspect, a method comprises: for each of a plurality of resource locators: generating first value for the resource locator that indicates, for a plurality of first sets of search results that each include the resource locator, a number of occurrences of other resource locators that were impressed and not selected when the resource locator was selected; generating a second value for the resource locator that indicates, for a plurality of second sets of search results that each include the resource locator, a number of occurrences of other resource locators that were selected when the resource locator was impressed and not selected; and generating, based on a difference between the first value and the second value, an adjustment factor for the resource locator for adjusting a score associated with the resource locator during a search operation.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for receiving audio data encoding an utterance; obtaining an intermediate transcription of the utterance; before a final transcription of the utterance is obtained: i) determining that the intermediate transcription of the utterance is associated with a previously received search query, ii) obtaining one or more results that are identified as responsive to the previously received search query, and iii) storing one or more of the results; obtaining the final transcription of the utterance; determining that the final transcription of the utterance is also associated with the previously received search query; and in response to determining that the final transcription of the utterance is also associated with the previously received search query, providing the stored one or more results for output.

Abstract: Methods, systems, and media for recommending media content based on attribute grouped viewing sessions are provided.

Abstract: Systems and methods are provided for dynamically displaying hyperlinks in a video based on various factors associated with a device at which the video is played and/or a user of the device. In one or more aspects, a system includes a request component configured to receive a request to play a video hosted by a media provider. The system further includes a selection component configured to select a subset of links included in a set of links associated with the video to provide with the video when it is played in response to the request based in part on a number of links included in the set of links, wherein graphical elements respectively representative of the links included in the subset of links are configured to be displayed over the video when the video is played in response to the request.

Abstract: Among other things, this document describes a computer-implemented method. The method can include receiving, at a first device, an indication of user input to cause the first device to establish a wireless data connection with another device. A determination can be made at the first device that one or more sensors on the first device are oriented toward a second device. In response to at least one of (i) receiving the indication of user input to cause the first device to establish a wireless data connection with another device and (ii) determining that the one or more sensors on the first device are oriented toward the second device, a first wireless data connection can be established between the first device and the second device. A first stream of audio data can be received and played at the first device.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating representation of acoustic sequences. One of the methods includes: receiving an acoustic sequence, the acoustic sequence comprising a respective acoustic feature representation at each of a plurality of time steps; processing the acoustic feature representation at an initial time step using an acoustic modeling neural network; for each subsequent time step of the plurality of time steps: receiving an output generated by the acoustic modeling neural network for a preceding time step, generating a modified input from the output generated by the acoustic modeling neural network for the preceding time step and the acoustic representation for the time step, and processing the modified input using the acoustic modeling neural network to generate an output for the time step; and generating a phoneme representation for the utterance from the outputs for each of the time steps.

Abstract: A system comprising at least one processor; and at least one storage device. The storage device(s) store instructions that, when executed, cause the at least one processor to: prior to enabling output of an audio signal based on an audio data stream, detect, within the audio data stream, an indication of a target sound that corresponds to one of a plurality of sounds that are expected to cause distraction, replace, within the audio data stream, the indication of the target sound with an indication of a replacement sound, wherein the replacement sound is a less distracting version of the target sound, and after replacing the indication of the target sound with the indication of the replacement sound, output the audio data stream.

Abstract: A user interface adaptation module identifies a dominant color of a portion selection of a frame of a video and, based on the dominant color, generates colors for components of a user interface in which the video is displayed. The colors of the user interface components are set based upon the generated colors and upon context information such as a playing state of the video. The setting of the component colors in this way allows the user interface to adjust to complement both the played content of the video and the video's context. In one embodiment, the dominant color is identified by partitioning individual pixels of the portion selection based on their respective colors. In one embodiment, a set of primary color variants is generated based on the dominant color, and different colors are generated for each type of user interface component based on the different primary color variants.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for neural translation systems with rare word processing. One of the methods is a method training a neural network translation system to track the source in source sentences of unknown words in target sentences, in a source language and a target language, respectively and includes deriving alignment data from a parallel corpus, the alignment data identifying, in each pair of source and target language sentences in the parallel corpus, aligned source and target words; annotating the sentences in the parallel corpus according to the alignment data and a rare word model to generate a training dataset of paired source and target language sentences; and training a neural network translation model on the training dataset.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for storing a plurality of items, each item including digital content, for each item of the plurality of items, generating a quality score to provide a plurality of quality scores, each quality score indicating a quality of an associated item and being based on at least one of a status score and a content score, the status score being associated with an author user of a respective item and the content score being associated with digital content provided in the respective item, determining an order of items based on respective quality scores, and transmitting instructions to display items to a user based on the order.

Abstract: An example system is described that includes a head-mountable device (HMD) with a bone conduction transducer that transmits an audio signal and an inertial measurement unit (IMU) sensor connected to the bone conduction transducer. The IMU sensor detects movement of the HMD due to vibrations of the bone conduction transducer transmitting the audio signal and the IMU sensor provides outputs. The system also includes one or more processors that receive the outputs of the IMU sensor and compare the outputs of the IMU sensor with at least one reference signal. The at least one reference signal is based on the audio signal that is transmitted. The one or more processors output an HMD fitting parameter based on the comparison.

Abstract: A method of cooling electronic equipment is disclosed and includes substantially continuously circulating ambient air across a plurality of rack-mounted electronic devices, monitoring the temperature of air in or around a group of devices in the plurality of rack-mounted electronic devices, and providing substantially cooler-than-ambient air to the group of devices when the monitored air temperatures rises above a set value.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, relating to generating log-linear models. In some implementations, n-gram parameter values derived from an n-gram language model are obtained. N-gram features for a log-linear language model are determined based on the n-grams corresponding to the obtained n-gram parameter values. A weight for each of the determined n-gram features is determined, where the weight is determined based on (i) an n-gram parameter value that is derived from the n-gram language model and that corresponds to a particular n-gram, and (ii) an n-gram parameter value that is derived from the n-gram language model and that corresponds to an n-gram that is a sub-sequence within the particular n-gram. A log-linear language model having the determined n-gram features is generated, where the determined n-gram features in the log-linear language model have weights that are initialized based on the determined weights.

Abstract: The present invention pertains to the allocation of bandwidth on a global large scale network. Systems and methods are provided to base bandwidth allocation on the predicted bandwidth demands of the network users. In allocating bandwidth, each user is assigned a user weight value, and is to be allocated bandwidth in proportion to that value. In addition, the bandwidth allocation is enforced so as to ensure that bandwidth allocation policies are maintained.