Abstract: This specification describes computer-implemented methods and systems. One method includes receiving, by a neural network of a speech recognition system, first data representing a first raw audio signal and second data representing a second raw audio signal. The first raw audio signal and the second raw audio signal describe audio occurring at a same period of time. The method further includes generating, by a spatial filtering layer of the neural network, a spatial filtered output using the first data and the second data, and generating, by a spectral filtering layer of the neural network, a spectral filtered output using the spatial filtered output. Generating the spectral filtered output comprises processing frequency-domain data representing the spatial filtered output. The method still further includes processing, by one or more additional layers of the neural network, the spectral filtered output to predict sub-word units encoded in both the first raw audio signal and the second raw audio signal.

Abstract: A method is disclosed. The method includes determining whether the counter has reached a threshold value. The method also includes sending a request to reset the counter, to a payment authorizing entity, if it is determined that the counter has reached the threshold value, and then receiving, from the payment authorizing entity, a script in response to the request.

Abstract: A system and method of establishing a resource provider as a trusted listing are disclosed. The method includes receiving, by a directory server computer, an indication from a user that a resource provider is trusted. The directory server computer is programmed to provide a first level of authentication. The method then includes storing, in a database, data representing the indication from the user that the resource provider is trusted. The method then includes receiving an authentication request message from the user conducting an interaction at the resource provider computer and determining that the data representing the indication from the user that the resource provider is trusted is present. In response to determining, the method includes providing a second level of authentication to the user before the user is allowed to complete the interaction. The second level of authentication is lower than the first level.

Abstract: A defense system that receives information regarding an incoming object(s), then automatically coordinates spoofing or jamming of SATNAV signals potentially used by the incoming object(s) while also informing friendly systems of the spoofing or jamming of SATNAV signal.

Abstract: This specification describes computer-implemented methods and systems. One method includes receiving, by a neural network of a speech recognition system, first data representing a first raw audio signal and second data representing a second raw audio signal. The first raw audio signal and the second raw audio signal describe audio occurring at a same period of time. The method further includes generating, by a spatial filtering layer of the neural network, a spatial filtered output using the first data and the second data, and generating, by a spectral filtering layer of the neural network, a spectral filtered output using the spatial filtered output. Generating the spectral filtered output comprises processing frequency-domain data representing the spatial filtered output. The method still further includes processing, by one or more additional layers of the neural network, the spectral filtered output to predict sub-word units encoded in both the first raw audio signal and the second raw audio signal.

Abstract: Devices, systems, and methods for controlling a contactless interface on a dual interface portable device are provided. The methods allow for contactless enablement and disablement on both a temporary and semi-permanent basis, dependent on commands sent to the device by an issuer of the portable device. The commands are received through a terminal device and are transferred to the portable device during a contact transaction. Configuration data on the portable device is modified according to the received commands. An application on the portable device controls the transmission of data through the contactless interface dependent on the configuration data.

Abstract: Systems, apparatus, and methods for managing an application installed on a mobile device, such as a mobile phone are provided, so as to prevent or limit unauthorized use of the application, for example, when the mobile device is lost, stolen, or otherwise subject to unauthorized use. A request to enable or disable user access to a function of the application can be received by a server computer. A command message responsive to the request can be generated by the server computer, the command message comprising control data for enabling or disabling user access to the function of the application. The generated command message can be transmitted to the mobile device over a wireless network.

Abstract: A system, apparatus, and method for processing payment transactions that are conducted using a mobile device that includes a contactless element, such as an integrated circuit chip. An account holder is enabled to generate transaction related data and append that data to a transaction record or transaction identifier. The appended data and transaction record or identifier may be stored in the mobile payment device and/or provided to an Issuer. If provided to an Issuer, the transaction related data generated by the account holder may be used to supplement an account statement. The additional information generated by the account holder may be used to assist the account holder in determining if a transaction is valid by providing information that helps the account holder to recall the location or other aspect of the transaction.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for processing audio waveforms. In some implementations, a time-frequency feature representation is generated based on audio data. The time-frequency feature representation is input to an acoustic model comprising a trained artificial neural network. The trained artificial neural network comprising a frequency convolution layer, a memory layer, and one or more hidden layers. An output that is based on output of the trained artificial neural network is received. A transcription is provided, where the transcription is determined based on the output of the acoustic model.

Abstract: A system, apparatus, and method for processing payment transactions that are conducted using a mobile device that includes a contactless element, such as an integrated circuit chip. The invention enables the updating, correction or synchronization of transaction data maintained by an Issuer with that stored on the device. This is accomplished by using a wireless (cellular) network as a data communication channel for data provided by an Issuer to the mobile device, and is particularly advantageous in circumstances in which the contactless element is not presently capable of communication with a device reader or point of sale terminal that uses a near field communications mechanism. Data transferred between the mobile device and Issuer may be encrypted and decrypted to provide additional security and protect the data from being accessed by other users or applications.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for audio-visual speech separation. A method includes: obtaining, for each frame in a stream of frames from a video in which faces of one or more speakers have been detected, a respective per-frame face embedding of the face of each speaker; processing, for each speaker, the per-frame face embeddings of the face of the speaker to generate visual features for the face of the speaker; obtaining a spectrogram of an audio soundtrack for the video; processing the spectrogram to generate an audio embedding for the audio soundtrack; combining the visual features for the one or more speakers and the audio embedding for the audio soundtrack to generate an audio-visual embedding for the video; determining a respective spectrogram mask for each of the one or more speakers; and determining a respective isolated speech spectrogram for each speaker.

Abstract: A system, apparatus, and method for processing payment transactions that are conducted using a mobile device that includes a contactless element, such as an integrated circuit chip. The invention enables the updating, correction or synchronization of transaction data maintained by an Issuer with that stored on the device. This is accomplished by using a wireless (cellular) network as a data communication channel for data provided by an Issuer to the mobile device, and is particularly advantageous in circumstances in which the contactless element is not presently capable of communication with a device reader or point of sale terminal that uses a near field communications mechanism. Data transferred between the mobile device and Issuer may be encrypted and decrypted to provide additional security and protect the data from being accessed by other users or applications.

Abstract: A defense system that receives information regarding an incoming object(s), then automatically coordinates spoofing or jamming of SATNAV signals potentially used by the incoming object(s) while also informing friendly systems of the spoofing or jamming of SATNAV signal.

Abstract: A system, apparatus, and method for processing payment transactions that are conducted using a mobile device that includes a contactless element, such as an integrated circuit chip. The invention enables the updating, correction or synchronization of transaction data maintained by an Issuer with that stored on the device. This is accomplished by using a wireless (cellular) network as a data communication channel for data provided by an Issuer to the mobile device, and is particularly advantageous in circumstances in which the contactless element is not presently capable of communication with a device reader or point of sale terminal that uses a near field communications mechanism. Data transferred between the mobile device and Issuer may be encrypted and decrypted to provide additional security and protect the data from being accessed by other users or applications.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for neural network adaptive beamforming for multichannel speech recognition are disclosed. In one aspect, a method includes the actions of receiving a first channel of audio data corresponding to an utterance and a second channel of audio data corresponding to the utterance. The actions further include generating a first set of filter parameters for a first filter based on the first channel of audio data and the second channel of audio data and a second set of filter parameters for a second filter based on the first channel of audio data and the second channel of audio data. The actions further include generating a single combined channel of audio data. The actions further include inputting the audio data to a neural network. The actions further include providing a transcription for the utterance.

Abstract: A counting system has a dispensing assembly, a singulator assembly, a counting assembly, and a control system. The dispensing assembly distributes particles to the singulator assembly which separates the particles into a singulator path and delivers the singulated particles to the counting assembly where the particles are counted and the count displayed on an input and display device.

Abstract: A defense system that receives information regarding an incoming object(s), then automatically coordinates spoofing or jamming of SATNAV signals potentially used by the incoming object(s) while also informing friendly systems of the spoofing or jamming of SATNAV signal.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for neural network adaptive beamforming for multichannel speech recognition are disclosed. In one aspect, a method includes the actions of receiving a first channel of audio data corresponding to an utterance and a second channel of audio data corresponding to the utterance. The actions further include generating a first set of filter parameters for a first filter based on the first channel of audio data and the second channel of audio data and a second set of filter parameters for a second filter based on the first channel of audio data and the second channel of audio data. The actions further include generating a single combined channel of audio data. The actions further include inputting the audio data to a neural network. The actions further include providing a transcription for the utterance.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for processing audio waveforms. In some implementations, a time-frequency feature representation is generated based on audio data. The time-frequency feature representation is input to an acoustic model comprising a trained artificial neural network. The trained artificial neural network comprising a frequency convolution layer, a memory layer, and one or more hidden layers. An output that is based on output of the trained artificial neural network is received. A transcription is provided, where the transcription is determined based on the output of the acoustic model.

Abstract: A portable consumer device may have money stored therein, which can be calculated using an accumulator record and an exception record. The accumulator record can be incremented by the amount of any initiated transaction using the portable consumer device. The exception record can be incremented by the amount of any completed transaction using the portable consumer device. The portable consumer device may also have a limit amount, to record the amount of money added. The records allow for a later reconstruction of the device history, and can account for any errors or problems during transactions.