Abstract: The present disclosure provides a method and apparatus for detecting a voice, relates to the fields of voice processing and deep learning technology. The method may include: acquiring a target voice; and inputting the target voice into a pre-trained deep neural network to obtain whether the target voice has a sub-voice in each of a plurality of preset direction intervals, the deep neural network being used to predict whether the voice has a sub-voice in each of the plurality of direction intervals.

Abstract: A method, apparatus, device, and storage medium for voice interaction. A specific embodiment of the method includes: extracting an acoustic feature from received voice data, the acoustic feature indicating a short-term amplitude spectrum characteristic of the voice data; applying the acoustic feature to a type recognition model to determine an intention type of the voice data, the intention type being one of an interaction intention type and a non-interaction intention type, and the type recognition model being constructed based on the acoustic feature of training voice data; and performing an interaction operation indicated by the voice data, based on determining that the intention type is the interaction intention type.

Abstract: The disclosure provides a speech recognition method, a device and a computer-readable storage medium. The method includes obtaining a first voice signal collected from a first microphone in a microphone array and a second voice signal collected from a second microphone in the microphone array, the microphone array including at least two microphones, such as two, three or six microphones. The method further includes extracting enhanced features associated with the first voice signal and the second voice signal through a neural network, and obtaining a speech recognition result based on the enhanced features extracted.

Abstract: Disclosed are systems, devices and methods for imaging and image-based sorting of particles in a flow system, including cells in a flow cytometer. In some aspects, a system includes a particle flow device to flow particles through a channel, an imaging system to obtain image data of a particle during flow through the channel, a processing unit to determine a property associated with the particle and to produce a control command for sorting the particle based on sorting criteria associated with particle properties, and an actuator to direct the particle into one of a plurality of output paths of the particle flow device in real-time.

Abstract: Apparatuses, methods, and systems for touch pad sensors that are further operative as antennas, are disclosed. One apparatus includes a plurality of touch sensors, a controller, and a radio. The controller is operative to receive sensed signals of the plurality of touch sensors, and provide controls of the apparatus in response to the receive sensed signals of the plurality of touch sensors. The radio is operative to wirelessly communicate with an external wireless device. One or more conductive lines electrically connect the radio with one or more of the plurality of touch sensors, wherein the plurality of touch sensors operate as one or more antennas for facilitating the wireless communication.

Abstract: Apparatuses, methods, and systems for a hinge feed antenna of eyewear, are disclosed. One apparatus includes a front frame, a temple arm, and a hinge rotatably connecting the front frame and the temple arm. For an embodiment, the temple arm includes a circuit assembly including at least a radio that supports wireless communication. For an embodiment, the front frame includes one or more antennas traces, wherein the hinge conductively connects the circuit assembly to the one or more antenna traces, and wherein the hinge in combination with the one or more antenna traces form one or more antennas for supporting the wireless communication.

Abstract: The present disclosure discloses an interactive voice-control method and apparatus, a device and a medium. The method includes: obtaining a sound signal at a voice interaction device and recognized information that is recognized from the sound signal; determining an interaction confidence of the sound signal based at least on at least one of an acoustic feature representation of the sound signal and a semantic feature representation associated with the recognized information; determining a matching status between the recognized information and the sound signal; and providing the interaction confidence and the matching status for controlling a response of the voice interaction device to the sound signal.

Abstract: Land cover (LC) and land use (LU) have commonly been classified separately from remotely sensed imagery, without considering the intrinsically hierarchical and nested relationships between them. A novel joint deep learning framework is proposed and demonstrated for LC and LU classification. The proposed Joint Deep Learning (JDL) model incorporates a multilayer perceptron (MLP) and convolutional neutral network (CNN), and is implemented via a Markov process involving iterative updating. In the JDL, LU classification conducted by the CNN is made conditional upon the LC probabilities predicted by the MLP. In turn, those LU probabilities together with the original imagery are re-used as inputs to the MLP to strengthen the spatial and spectral feature representation. This process of updating the MLP and CNN forms a joint distribution, where both LC and LU are classified simultaneously through iteration.

Abstract: An object-based convolutional neural network (OCNN) method and system for urban land use classification from VFSR imagery are described. In the OCNN, segmented objects consisting of linearly shaped objects (LS-objects) and other general objects (G-objects), are utilized as functional units. The G-objects are precisely identified and labelled through a single large input window (128×128) CNN with a deep (eight-layer) network to perform a contextual object-based classification. Whereas the LS-objects are each distinguished accurately using a range of small input window (48×48) CNNs with less deep (six-layer) networks along the objects' lengths through majority voting. The locations of the input image patches for both CNN networks are determined by considering both object geometry and its spatial anisotropy, such as to accurately classify the objects into urban land use classes.

Abstract: The present invention relates to methods for a blockchain search system, and particularly, the present invention relates to methods for assuring the data integrity on blockchain hybrid storage. Furthermore, the present invention provides a cost-efficient update scheme to maintain the on-chain index and authenticated data structure and is efficient and scalable under various operational parameters.

Abstract: An object-based convolutional neural network (OCNN) method and system for urban land use classification from VFSR imagery are described. In the OCNN, segmented objects consisting of linearly shaped objects (LS-objects) and other general objects (G-objects), are utilized as functional units. The G-objects are precisely identified and labelled through a single large input window (128×128) CNN with a deep (eight-layer) network to perform a contextual object-based classification. Whereas the LS-objects are each distinguished accurately using a range of small input window (48×48) CNNs with less deep (six-layer) networks along the objects' lengths through majority voting. The locations of the input image patches for both CNN networks are determined by considering both object geometry and its spatial anisotropy, such as to accurately classify the objects into urban land use classes.

Abstract: Land cover (LC) and land use (LU) have commonly been classified separately from remotely sensed imagery, without considering the intrinsically hierarchical and nested relationships between them. A novel joint deep learning framework is proposed and demonstrated for LC and LU classification. The proposed Joint Deep Learning (JDL) model incorporates a multilayer perceptron (MLP) and convolutional neutral network (CNN), and is implemented via a Markov process involving iterative updating. In the JDL, LU classification conducted by the CNN is made conditional upon the LC probabilities predicted by the MLP. In turn, those LU probabilities together with the original imagery are re-used as inputs to the MLP to strengthen the spatial and spectral feature representation. This process of updating the MLP and CNN forms a joint distribution, where both LC and LU are classified simultaneously through iteration.

Abstract: Disclosed are systems, devices and methods for imaging and image-based sorting of particles in a flow system, including cells in a flow cytometer. In some aspects, a system includes a particle flow device to flow particles through a channel, an imaging system to obtain image data of a particle during flow through the channel, a processing unit to determine a property associated with the particle and to produce a control command for sorting the particle based on sorting criteria associated with particle properties, and an actuator to direct the particle into one of a plurality of output paths of the particle flow device in real-time.

Abstract: A planar multi-layer assembly method fabricates a dual frequency, dual polarization phased-array antenna. A plurality of vias make up an array of double-walled wells which are connected to a ground plane. A shorted annular ring patch antenna (SARPA) is deposited at the top of each double-walled well. Fabricated coaxially and parallel to each SARPA, is an array of circular patch antennas (CPA). The inner wall of each double-walled well improves isolation of the CPA signals from the SARPA signals. Each SARPA of the array is connected to a pair of first frequency band signal vias and the CPA is coupled to a pair of second frequency band signal vias. Within each frequency band, a plurality of signal phases enable steerable polarized antenna beams.

Abstract: An electromagnetic valve includes a valve body, a valve core, and an electromagnetic drive device. The valve body has an operation chamber, wherein the valve core is movably arranged in the operation chamber. The electromagnetic drive device is arranged in the valve core and is suitable for being electrically connected with a power source. The electromagnetic drive device is arranged, when driving the electromagnetic valve, to provide the valve core with a starting drive so as to drive the valve core to enter a running state. After the valve core is driven to enter the running to state, to provide the valve core with a maintaining drive so as to keep the valve core in the running state.

Abstract: A method includes receiving, by a first wireless device, first data indicative of channel properties of a first communication link between the first wireless device and a second wireless device, the first wireless device and the second wireless device being located in a same vicinity. The method further includes detecting, using a supervised machine learning (ML) model applied to the first data, human movement and presence of stationary objects within the same vicinity and activating, by the first wireless device in response to detection of the human movement, an ultrasonic signal. The method further includes receiving, by the first wireless device, a reflected component of the ultrasonic signal and confirming, by the first wireless device using the reflected component, presence of a human in the same vicinity.

Abstract: A system and method include processing logic receiving, from a wireless transceiver of a first device, first data indicative of channel state information (CSI) of a first communication link between the wireless transceiver and a wireless transmitter of a second device, the first device and the second device being located in a building. The logic pre-preprocesses the first data to generate input vectors composed of statistical parameter values derived from sets of discrete samples of the first data. The logic processes, through a long short-term memory (LSTM) layer of a neural network, the input vectors to generate multiple hidden state values of the neural network. The logic processes, through a set of additional layers of the neural network, respective hidden state values of the multiple hidden state values to determine that a human is present in the building.

Abstract: A planar multi-layer assembly method fabricates a dual frequency, dual polarization phased-array antenna. A plurality of vias make up an array of double-walled wells which are connected to a ground plane. A shorted annular ring patch antenna (SARPA) is deposited at the top of each double-walled well. Fabricated coaxially and parallel to each SARPA, is an array of circular patch antennas (CPA). The inner wall of each double-walled well improves isolation of the CPA signals from the SARPA signals. Each SARPA of the array is connected to a pair of first frequency band signal vias and the CPA is coupled to a pair of second frequency band signal vias. Within each frequency band, a plurality of signal phases enable steerable polarized antenna beams.

Abstract: In a disclosed method, a computing device receiver, from a wireless receiver (RX), first data indicative of channel properties of a first communication link between the wireless receiver (RX) in a first device and a wireless transmitter (TX) in a second device. The first device and the second device are located in a building. The computing device further executes a neural network to process the first data to distinguish humans from stationary objects within the building and detect presence of the human in the building. The computing device transmits result data indicative of the presence to at least one of the first device or the second device.

Abstract: A planar multi-layer assembly method fabricates a dual frequency, dual polarization phased-array antenna. A plurality of vias make up an array of double-walled wells which are connected to a ground plane. A shorted annular ring patch antenna (SARPA) is deposited at the top of each double-walled well. Fabricated coaxially and parallel to each SARPA, is an array of circular patch antennas (CPA). The inner wall of each double-walled well improves isolation of the CPA signals from the SARPA signals. Each SARPA of the array is connected to a pair of first frequency band signal vias and the CPA is coupled to a pair of second frequency band signal vias. Within each frequency band, a plurality of signal phases enable steerable polarized antenna beams.