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: A computerized method of managing a computer remote session operation, comprising providing a server for hosting application execution; configuring a number of predefined user accounts with low security permissions on said server, where said user accounts are not tied to any specific real user; Whenever a remote user requests to start a remote session, finding an available user account not currently in use on said computer, allocating it for the remote session and marking it as unavailable for subsequent session requests; Generating a one-time password for said user account; Communicating the assigned user account identifier and temporary password to client component on the user's side, either directly or through an intermediate broker; causing the client component to connect to the server using said user account identifier and temporary password; and, upon termination of the remote session, deleting the assigned user account's data and marking it as available again.

Abstract: This application is directed to changing a TV operating mode based on a potential relationship among viewers of a connected TV. In some implementations, a method includes: determining viewer identities and identifying the potential relationship among the viewers or a subset of the viewers; in accordance with a determination that the potential relationship allows the viewers to share private information, placing or maintaining the connected TV in or into a first operating mode that is unrelated to media content being displayed and allows personal information to be displayed on the connected TV; and in accordance with a determination that the potential relationship does not allow the viewers to share private information, placing or maintaining the connected TV in or into a second operating mode that is unrelated to the media content and does not allow personal information to be displayed on the connected TV.

Abstract: A processor of a computing device is described that intercepts a write command output from a file system, the write command specifying new data to be written to a particular group of blocks of a storage device. The processor determines, based on a record of free blocks of the storage device, that the particular group of blocks of the storage device include one or more free blocks that are currently storing snapshot data as old data. Responsive to determining that the particular group of blocks of the storage device include one or more free blocks that are currently storing snapshot data as old data, the processor moves the old data to other free blocks of the storage device, and after moving the old data to the other free blocks, the processor forwards the write command to the storage device to write the new data to the particular group of blocks.

Abstract: This application is directed to an electronic device powered by one or more rechargeable battery cells. The electronic device includes a first negative temperature coefficient (NTC) thermistor proximate to the battery cells, and an open capacitor coupled in parallel with the NTC thermistor. The open capacitor has an open area and two electrodes that are at least partially exposed via the open area and electrically isolated. The electronic device further includes a control circuit coupled to the NTC thermistor and the open capacitor. The control circuit is configured to detect a voltage drop across the NTC thermistor and the open capacitor if conductive liquid enters the open area of the capacitor and electrically connects the two electrodes that are at least partially exposed via the open area.

Abstract: Techniques to provide uniform luminance across a computing device display, such as an active matrix organic light emitting diode (AMOLED) display. In some examples, a computing device display may include a hole within the active area of the display that may be used for a camera, a button or some other function. The hole may result in a non-uniform voltage drop in a power supply mesh in regions of the active area near the hole. The power supply mesh may provide electrical energy to elements of the display. The techniques of this disclosure include portions of the power supply mesh that are not connected to the voltage supply bus to ensure a uniform voltage drop across the active area of the display.

Abstract: A method for Short-Time Fourier Transform-based echo muting includes receiving a microphone signal including acoustic echo captured by a microphone and corresponding to audio content from an acoustic speaker, and receiving a reference signal including a sequence of frames representing the audio content. For each frame in a sequence of frames, the method includes processing, using an acoustic echo canceler configured to receive a respective frame as input to generate a respective output signal frame that cancels the acoustic echo from the respective frame, and determining, using a Double-talk Detector (DTD), based on the respective frame and the respective output signal frame, whether the respective frame includes a double-talk frame or an echo-only frame. For each respective frame that includes the echo-only frame, muting the respective output signal frame, and performing speech processing on the respective output signal frame for each respective frame that includes the double-talk frame.

Abstract: A method (800) to detect a hotword in a spoken utterance (120) includes receiving a sequence of input frames (210) characterizing streaming multi-channel audio (118). Each channel (119) of the streaming multi-channel audio includes respective audio features (510) captured by a separate dedicated microphone (107). For each input frame, the method includes processing, using a three-dimensional (3D) single value decomposition filter (SVDF) input layer (302) of a memorized neural network (300), the respective audio features of each channel in parallel and generating a corresponding multi-channel audio feature representation (420) based on a concatenation of the respective audio features (344). The method also includes generating, using sequentially-stacked SVDF layers (350), a probability score (360) indicating a presence of a hotword in the audio. The method also includes determining whether the probability score satisfies a threshold and, when satisfied, initiating a wake-up process on a user device (102).

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for speaker diarization are disclosed. In one aspect, a method includes the actions of receiving audio data corresponding to an utterance. The actions further include determining that the audio data includes an utterance of a predefined hotword spoken by a first speaker. The actions further include identifying a first portion of the audio data that includes speech from the first speaker. The actions further include identifying a second portion of the audio data that includes speech from a second, different speaker. The actions further include transmitting the first portion of the audio data that includes speech from the first speaker and suppressing transmission of the second portion of the audio data that includes speech from the second, different speaker.

Abstract: The technology relates to an integrated circuit (IC) package in which an interconnection interface chiplet and/or interconnection interface circuit are relocated, partitioned, and/or decoupled from a main or core IC die and/or high-bandwidth memory (HBM) components in an integrated component package.

Abstract: A method includes identifying a first table including data. The first table has associated metadata, an associated replication state, an associated replication log file including replication logs logging mutations of the first table, and an associated replication configuration file including a first association that associates the first table with a replication family. The method includes inserting a second association in the replication configuration file that associates a second table having a non-loadable state with the replication family. The association of the second table with the replication family causes persistence of any replication logs in the replication log file that correspond to any mutations of the first table during the existence of the second table. The method further includes generating a third table from the first table, the metadata associated with the first table, and the associated replication state of the first table.

Abstract: A wireless communication device identifies an access point for a wireless network. The device initiates a connection procedure to establish a wireless connection between the wireless communication device and the access point, and the connection procedure includes: (i) selecting a degraded set of uplink transmission quality parameters that correspond to a higher signal-to-noise (SNR) ratio than a baseline set of uplink transmission quality parameters, and (ii) transmitting messages from the wireless communication device to the access point during the connection procedure according to the degraded set of uplink transmission quality parameters. The wireless network connection can be established between the wireless communication device and the access point as a result of completion of the connection procedure, despite the wireless communication device transmitting messages to the access point during the connection procedure according to the degraded set of uplink transmission quality parameters.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for designating certain voice commands as hotwords. The methods, systems, and apparatus include actions of receiving a hotword followed by a voice command. Additional actions include determining that the voice command satisfies one or more predetermined criteria associated with designating the voice command as a hotword, where a voice command that is designated as a hotword is treated as a voice input regardless of whether the voice command is preceded by another hotword. Further actions include, in response to determining that the voice command satisfies one or more predetermined criteria associated with designating the voice command as a hotword, designating the voice command as a hotword.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing a machine learning task on a network input to generate a network output. One of the systems comprises an attention neural network configured to perform the machine learning task, the attention neural network comprising a plurality of attention layers, each attention layer comprising an attention sub-layer that is arranged in parallel with a feed-forward sub-layer.

Abstract: A mobile computing device includes one or more processors, an emissive display that includes a touchscreen, and a proximity sensor that generates a quantitative output signal. The one or more processors transition, in response to the quantitative output signal decreasing below a low threshold value, operation of the sensor to a high sensitivity state; transition, in response to the quantitative output signal increasing above a high threshold value, operation of the sensor to the low sensitivity state; and deactivate the touchscreen and/or the emissive display when the touchscreen and/or the emissive display is activated and when the quantitative output signal increases above an high threshold value and to activate the touchscreen and/or the emissive display when the touchscreen and/or the emissive display is deactivated and when the quantitative output signal is below a low threshold value, the low threshold value being less than the high threshold value.

Abstract: In general, the subject matter described in this disclosure can be embodied in methods, devices, and program products for modifying interactions between a mobile device and a peripheral device based on proximity. The mobile device monitors a wireless communication channel and receives a transmission from the peripheral device over the channel indicating that the mobile device is located within a first zone of proximity to the peripheral device. The mobile device determines distance values between the devices. When the mobile device determines that a distance value satisfies a threshold value, the mobile device visually indicates that it is within a second zone of proximity to the peripheral device. As a result of the mobile device receiving data from the peripheral device using a second communication protocol, the mobile device determines that it is located within a third zone of proximity to the peripheral device, and performs an action.

Abstract: A battery pack includes a battery, a first temperature sensor configured to provide a first temperature value associated with a temperature of the battery, a heat source disposed proximate to the battery and configured to heat the battery, a second temperature sensor configured to provide a second temperature value associated with a temperature of the heat source, and a control board coupled to the first temperature sensor and the second temperature sensor, wherein the control board is configured to receive the first temperature value and the second temperature value. The control board is configured to compare the first temperature value and the second temperature value to determine a temperature gradient between the battery and the heat source and transmit an alert if the temperature gradient exceeds a first temperature gradient threshold.

Abstract: Disclosed are devices, systems, and methods for operating a backup power source or an uninterruptible power supply (UPS) that can be used in data centers and that provide a backup power source to power the data center when utility power is compromised. A power delivery system that provides power to a primary system may include a UPS with a state timing control system that operates a bypass static switch. The state timing control system can determine when to transition the primary system from the utility power supply to the backup UPS, based on the current AC voltage conditions. The state timing control system may perform modeling to emulate the intermediate DC voltage of an actual rectifier, and particularly emulate the holdup capacitor voltage. The emulated capacitor voltage can be obtained in real time by both an input power model based on RMS utility voltage and the actual rectifier output load.

Abstract: Provided are methods, systems, and apparatuses for detecting, processing, and responding to audio signals, including speech signals, within a designated area or space. A platform for multiple media devices connected via a network is configured to process speech, such as voice commands, detected at the media devices, and respond to the detected speech by causing the media devices to simultaneously perform one or more requested actions. The platform is capable of scoring the quality of a speech request, handling speech requests from multiple end points of the platform using a centralized processing approach, a de-centralized processing approach, or a combination thereof, and also manipulating partial processing of speech requests from multiple end points into a coherent whole when necessary.

Abstract: A situation awareness system is provided and includes sensors configured to sense an object and generate signals of the sensed object. An identification and classification module is in communication with the sensors. The identification and classification module is configured to identify the sensed object and determine the position, speed and direction of each sensed object. A sound synthesis module is in communication with the identification and classification module. The sound synthesis module is configured to create sounds for the sensed object and further configured to create a binaural virtual acoustic model for the sensed object. The binaural virtual acoustic model is used to position the sensed object in virtual space at a location corresponding to its location in real space. One or more beam forming arrays is in communication with the sound synthesis module and configured to delivery sound. The delivered sound increases the situational awareness of a vehicle operator.