Abstract: A system and method for allowing a single live customer service agent to simultaneously serve multiple customers. According to various embodiments, a virtual agent operates at the front end to receive and attempt to handle customer issues. The virtual agent employs speech recognition and intent mapping in order to generate a proposed response that attempts to identify and resolve customer issues. In some scenarios, the proposed response includes both a response message and a response action to be taken. A chat history and the proposed response is then provided to the live agent. The live agent reviews the information provided, and determines whether the proposed response is appropriate. The live agent then approves the response or revises or replaces the response. The final response is sent back to the virtual agent for processing and providing to the customer.

Abstract: A method with object tracking includes: determining a first target tracking state by tracking a target from a first image frame with a first field of view (FoV); determining a second FoV based on the first FoV and the first target tracking state; and generating a second target tracking result by tracking a target from a second image frame with the second FoV.

Abstract: This application describes techniques for displaying products in livestreaming. The techniques comprise displaying a first interface in response to a display request triggered by operating a preset icon on a web-based livestreaming page, wherein the first interface comprises product information and at least one view button corresponding to at least one product; and displaying a second interface in response to a view request triggered by operating the at least one view button, wherein the second interface comprises an identification code, wherein the identification code is configured to be scanned by a mobile terminal and to enter an application-based livestreaming page corresponding to the web-based livestreaming page, wherein the application-based livestreaming page and the web-based livestreaming page are associated with a live room of a same online streamer, and wherein the application-based livestreaming page comprises detailed information of a product.

Abstract: In some embodiments, the present disclosure provides an exemplary method that may include steps of monitoring, a conversation script between a call center agent and a customer; utilizing, a speech-to-text deep machine learning model to transcribe the audio call to text; utilizing, a natural language processing deep machine learning model to map intent mappings of the audio call; utilizing, a similarity measurement model to determine a semantic similarity between predefined intent mappings and the intent mappings of the call audio text; determining, an error based on the semantic similarity in the intent mapping call audio text; determining a training session based on the error.

Abstract: System and methods are provided for dynamically parallelized large data set processing. A workflow includes a distributed map state. A set of steps in the distributed map state are run for each item in a collection of items or for each batch. Execution of the distributed map state is dynamically parallelized such that the system creates as many parallel workflows as there are items in the collection at run time (or batches). The items in the collection are dynamically determined by the workflow system, as defined by a resource in the distributed map state.

Abstract: In some implementations, a system may capture audio from a call between a calling device and a called device. The system may filter the captured audio to generate a background audio layer. The system may generate an audio footprint that is a representation of sound in the background audio layer. The system may determine that the audio footprint includes a triggering sound footprint based on one or more audio characteristics of the audio footprint. The system may detect synthetic sound based on the audio footprint and after determining that the audio footprint includes the triggering sound footprint, wherein the synthetic sound is indicative of a sound recording. The system may transmit a notification to one or more devices associated with the call based on detecting the synthetic sound.

Abstract: A molecular marker related to body weight traits of Hu sheep and an application are provided. The whole genome sequencing of Hu sheep is carried out in the invention, the combination of QTL and GWAS is adopted, the candidate genes related to the weight traits are analyzed and the three molecular markers of MAP3K1, ABCB1, and MEF2C are selected as candidate genes for the weight traits of Hu sheep, and the two molecular markers of ANKRD55 and TRNAW-CCA-87 can be used as potential candidate genes for the weight traits of Hu sheep, these candidate genes can be used as molecular markers for application, the guidance of Hu sheep breeding will help to understand the genetic mechanism of weight traits of Hu sheep and further guide Hu sheep breeding.

Abstract: A system may include processor(s), and memory in communication with the processor(s) and storing instructions configured to cause the system to correct ASR errors. The system may receive a transcription comprising transcribed word(s) and may determine whether the transcribed word(s) exceed associated predefined confidence level(s). Responsive to determining a transcribed word does not exceed a predefined confidence level, the system may generate a predicted word. The system may calculate a distance between numerical representations of the transcribed word and the predicted word and may determine whether the distance exceeds a predefined threshold. Responsive to determining the distance exceeds the predefined threshold, the system may determine whether at least one red flag word of a list of red flag words corresponds to a context of the transcription, and, responsive to making that determination, may classify the transcription as associated with a first category.

Abstract: A method with object tracking includes: determining a first target tracking state by tracking a target from a first image frame with a first field of view (FoV); determining a second FoV based on the first FoV and the first target tracking state; and generating a second target tracking result by tracking a target from a second image frame with the second FoV.

Abstract: In some implementations, a system may capture audio from a call between a calling device and a called device. The system may filter the captured audio to generate a background audio layer. The system may generate an audio footprint that is a representation of sound in the background audio layer. The system may determine that the audio footprint includes a triggering sound footprint based on one or more audio characteristics of the audio footprint. The system may detect synthetic sound based on the audio footprint and after determining that the audio footprint includes the triggering sound footprint, wherein the synthetic sound is indicative of a sound recording. The system may transmit a notification to one or more devices associated with the call based on detecting the synthetic sound.

Abstract: In some embodiments, the present disclosure provides an exemplary method that may include steps of monitoring, a conversation script between a call center agent and a customer; utilizing, a speech-to-text deep machine learning model to transcribe the audio call to text; utilizing, a natural language processing deep machine learning model to map intent mappings of the audio call; utilizing, a similarity measurement model to determine a semantic similarity between predefined intent mappings and the intent mappings of the call audio text; determining, an error based on the semantic similarity in the intent mapping call audio text; determining a training session based on the error.

Abstract: A system for generating contrast reformation data includes image processing circuitry configured to generate two-dimensional reformation data for three-dimensional image data by straightening a centerline of the lumen of an imaged patient and unfolding the three-dimensional medical image data about the straightened centerline. Points in the three-dimensional image data are registered to points in two-dimensional contrast image data. The unfolding is used to translate the registration for the three-dimensional image data to the two-dimensional reformation data. Contrast information from the two-dimensional contrast image data is super-imposed onto the two-dimensional reformation data to generate the contrast reformation data.

Abstract: A method for calling hardware across devices includes: in response to a trigger operation by a user for a virtual hardware service corresponding to target hardware, acquiring, through a hardware coordination service interface, hardware metadata of the target hardware; sending, through a hardware cross-device call interface, a hardware call instruction to the target hardware based on the hardware metadata; sending, through a data stream transmission interface, a data stream acquisition request for the target hardware to the called device; and receiving data stream information returned by the called device in response to the data stream acquisition request, where the data stream information is generated in response to the target hardware executing the hardware call instruction.

Abstract: Systems and computer-implemented methods disclosed herein relate to detecting errors in manually entered data. In one embodiment, the system can identify a named entity automatically from a conversation between a customer and service agent with a named entity recognition model that employs natural language processing and machine learning to detect a word or string of words in the conversation that corresponds to a named entity category. In another embodiment, the system can determine whether data entered into a field on a service platform by the service agent includes an error by comparing the data entered with the named entity. In another embodiment, the system can transmit an alert to the service agent through the service platform when there is a mismatch between the named entity and the data entered.

Abstract: An object emotion analysis method and apparatus and an electronic device are provided. The method includes: extracting a static facial feature and a dynamic feature from multimedia data associated with a target object, wherein the dynamic feature includes one or more of an expression change feature, a sound feature and a language content feature: inputting the static facial feature and the dynamic feature into a pre-trained object emotion analysis model, fusing the static facial feature and the dynamic feature by the object emotion analysis model, and outputting an emotion analysis result.

Abstract: In some embodiments, the present disclosure provides an exemplary method that may include steps of monitoring, a conversation script between a call center agent and a customer; utilizing, a speech-to-text deep machine learning model to transcribe the audio call to text; utilizing, a natural language processing deep machine learning model to map intent mappings of the audio call; utilizing, a similarity measurement model to determine a semantic similarity between predefined intent mappings and the intent mappings of the call audio text; determining, an error based on the semantic similarity in the intent mapping call audio text; determining a training session based on the error.

Abstract: A charging control method and an electronic device implementing a charging control method includes: sending a first data packet, in which, the first data packet includes to-be-charged identification information and first characteristic information of the first electronic device; receiving a second data packet corresponding to the first data packet sent by the second electronic device, in which, the second data packet includes power supply identification information and charging strategy information; determining and performing a charging strategy according to the second data packet, in which, the charging strategy includes at least one of: the first electronic device receiving power supply from the second electronic device, and the first electronic device supplying power to the second electronic device.

Abstract: A system and method for allowing a single live customer service agent to simultaneously serve multiple customers. According to various embodiments, a virtual agent operates at the front end to receive and attempt to handle customer issues. The virtual agent employs speech recognition and intent mapping in order to generate a proposed response that attempts to identify and resolve customer issues. In some scenarios, the proposed response includes both a response message and a response action to be taken. A chat history and the proposed response is then provided to the live agent. The live agent reviews the information provided, and determines whether the proposed response is appropriate. The live agent then approves the response or revises or replaces the response. The final response is sent back to the virtual agent for processing and providing to the customer.

Abstract: Provided herein is an apparatus for absorbing, capturing, and desorbing CO2 using solid amines, the apparatus comprising: a catcher used for capturing CO2 in flue gas and comprising a treatment chamber and a delivery pipe for adsorption of CO2 in the flue gas by means of a spherical solid amine adsorbent, the bottom end of the delivery pipe being provided with a flue gas inlet, a first solid amine inlet, and an adsorbent inlet; a desorber for heating and desorbing CO2 adsorbed by the solid amine adsorbent; and a cooler for cooling the solid amine adsorbent from the desorber. Also provided is a method for absorbing, capturing and desorbing CO2 using solid amine. The present invention realizes continuous operation of capture and desorption and can be used for capturing CO2 in large-scale industrial flue gas.

Abstract: A medical image associated with a volumetric medical dataset including a sheared volume, which dataset is received from a medical scanner, is rendered. The sheared volume is a series of 2D slices and a sheared direction. An auxiliary rectilinear volume is constructed such that the 2D slices are preserved, albeit with displaced 2D lattice, and the sheared direction is replaced by a direction orthogonal to the 2D slices. Nearest points on the sheared and the auxiliary rectilinear 3D lattices are determined. A value of a (e.g., binary) scalar field at a point on the sheared 3D lattice is assigned to the nearest point on the auxiliary rectilinear 3D lattice, where a Euclidean distance transform (EDT) is applied to the value of the scalar field. Based on the EDT value of the scalar field, the medical image is rendered.