Abstract: A voice identity feature extractor training method includes extracting a voice feature vector of training voice, The method may include determining a corresponding I-vector according to the voice feature vector of the training voice. The method may include adjusting a weight of a neural network model by using the I-vector as a first target output of the neural network model, to obtain a first neural network model, The method may include obtaining a voice feature vector of target detecting voice and determining an output result of the first neural network model for the voice feature vector of the target detecting voice. The method may include determining an I-vector latent variable. The method may include estimating a posterior mean of the I-vector latent variable, and adjusting a weight of the first neural network model using the posterior mean as a second target output, to obtain a voice identity feature extractor.

Abstract: This application relates to a method and an apparatus for controlling an object in a virtual scene. The method may include generating a first virtual object at a first position point outside a navigation surface in a virtual scene; determining an appearance action of the first virtual object at a second position point in the navigation surface; controlling the first virtual object to move from the first position point to the second position point; and controlling the first virtual object to perform the appearance action at the second position point. In this way, when a virtual object needs to be refreshed, the first virtual object may be generated at the first position point, the first virtual object is controlled to move to the second position point, and the appearance action is performed at the second position point, thereby improving a refreshing effect of the virtual object.

Abstract: A method, apparatus, and a storage medium for displaying a virtual prop includes: displaying a screen in which a target virtual operation object is located in a target application; detecting a virtual prop of a target type within a visual angle range of the target virtual operation object in response to the target virtual operation object being configured with a target virtual skill, where the target virtual skill allows the target virtual operation object to see the virtual prop of the target type through a virtual obstacle; and displaying perspective information of the target virtual prop in the target application in response to the target virtual prop of the target type being detected within the visual angle range of the target virtual operation object.

Abstract: Embodiments of this disclosure provide a speech signal processing method and apparatus, an electronic device, and a non-transitory computer-readable storage medium. A processing circuitry of the electronic device obtains speech feature parameters of a to-be-processed speech signal. The speech feature parameters include an initial spectrum of the to-be-processed speech signal. The processing circuitry predicts a first quantity of subband spectrum envelopes based on the speech feature parameters. The first set quantity is less than a quantity of initial spectrum coefficients included in the initial spectrum. Each of the subband spectrum envelopes corresponds to a second quantity of initial spectrum coefficients. The processing circuitry adjusts the initial spectrum based on the predicted subband spectrum envelopes to obtain an adjusted spectrum. The processing circuitry obtains an enhanced speech signal corresponding to the to-be-processed speech signal based on the adjusted spectrum.

Abstract: A method of living body detection includes generating a detection interface in response to a receipt of a living body detection request for a verification of a detection object. The detection interface includes a first region with a viewing target for the detection object to track, a position of the first region in the detection interface changes during a detection time according to a first sequence of a position change of the first region. The method also includes receiving a video stream that is captured during the detection time, determining a second sequence of a sight line change of the detection object based on the video stream, and determining that the detection object is a living body at least partially based on the second sequence of the sight line change of the detection object matching the first sequence of the position change of the first region.

Abstract: This application relates to a speech recognition method and apparatus, and a computer-readable storage medium, and the method includes: obtaining a first loss function of a speech separation and enhancement model and a second loss function of a speech recognition model; performing back propagation based on the second loss function to train an intermediate model bridged between the speech separation and enhancement model and the speech recognition model, to obtain a representation model; fusing the first loss function and the second loss function, to obtain a target loss function; and jointly training the speech separation and enhancement model, the representation model, and the speech recognition model based on the target loss function, and ending the training when a preset convergence condition is met.

Abstract: A processing circuitry obtains a training set based on an interaction process of a plurality of cards. The training set includes a plurality of video frames, and a video frame includes a trigger region for triggering an action during the interaction process. The processing circuitry determines feature regions in the trigger region. A feature region includes a card and is set with an action label for indicating a first training content in the feature region. The processing circuitry also determines a feature vector based on the plurality of video frames. The feature vector indicates a triggering of the feature region. The processing circuit inputs the feature vector into a first model for training to obtain a second model. The first model associates the feature vector with the action label, and the second model is used for indicating a correspondence between a content of the card and the action label.

Abstract: Aspects of the disclosure provide methods and apparatuses of virtual prop allocation and provide non-transitory computer-readable storage mediums storing instructions for virtual prop allocation. A method of virtual prop allocation includes obtaining a geographic location and state information of at least one terminal device participating in a first scene, and determining a first virtual prop associated with the first scene based on the geographic location and the state information of the at least one terminal device participating in the first scene. Further, the method includes determining geographic location information for releasing the first virtual prop, and transmitting, to a first terminal device in the at least one terminal device, the geographic location information for releasing the first virtual prop.

Abstract: A method for processing a frequency control signal includes providing a square wave pulse to a target qubit, and controlling, after a first time elapses from an end time of the square wave pulse, the target qubit to rotate around a Y axis by a first target distance. The first time has a value that is variable. The method includes performing, after a second time elapses from the first time, a QST measurement on the target qubit and reconstructing a density matrix of the target qubit based on the QST measurement to obtain a phase parameter value of the target qubit associated with the value of the first time. Further, the method includes varying the first time and repeating the QST measurement in response to values of the first time to obtain phase parameter values associated with the values of the first time; and adjusting the frequency control signal accordingly.

Abstract: A communication method of an edge computing platform in which an application function is deployed is provided. The edge computing platform is connected to a network exposure function-lite. The network exposure function-lite is connected to a network exposure function-central deployed in a core network. In the communication method, a capability invocation request is received from the application function. When the capability invocation request corresponds to a request to acquire wireless network information, a subscription request for the wireless network information is transmitted to an access network side based on the capability invocation request. The wireless network information fed back by the access network side is received in response to the subscription request. Further, the application function is notified of the wireless network information.

Abstract: An image segmentation method includes obtaining target domain images and source domain images, and segmenting the source domain images and the target domain images by using a generative network in a first generative adversarial network. The method further includes segmenting the source domain images and the target domain images by using a generative network in a second generative adversarial network, and determining a first source domain image and a second source domain image according to source domain segmentation losses, and determining a first target domain image and a second target domain image according to target domain segmentation losses. The method also includes performing cross training on the first generative adversarial network and the second generative adversarial network to obtain a trained first generative adversarial network; and segmenting a to-be-segmented image based on the generative network in the trained first generative adversarial network.

Abstract: An interface display update method is performed by an electronic device, including: displaying a virtual item currently controlled by a virtual object in a virtual scene; detecting a perspective rotation operation associated with the virtual object; obtaining a target rotation angle, a first rotation speed of a perspective of the virtual object, and a second rotation speed of the virtual item according to the perspective rotation operation, the second rotation speed being less than the first rotation speed; and updating the virtual scene, based on the target rotation angle, the first rotation speed of the perspective of the virtual object, and the second rotation speed of the virtual item, using a target animation in which the virtual scene changes in accordance with the perspective rotation operation, the target animation visualizing an effect that rotation of the virtual item lags behind rotation of the perspective of the virtual object.

Abstract: A method for dynamic encryption and signing includes: obtaining, by a terminal during session connection, a predetermined first key index and a randomly-generated first signature index; signing session request data with a first signature corresponding to the first signature index; encrypting the session request data with a first key corresponding to the first key index; sending the encrypted session request data and the first signature index to the server; and receiving, from the server, session response data signed with a second signature corresponding to a second signature index and encrypted with a second key corresponding to a second key index, after decryption and signature verification by the server over the session request data succeed. The second signature index and the second key index are randomly selected by the server and saved in a login session object accessible by both the server and the terminal.

Abstract: The embodiments of this application relate to a video encoding method performed at a computer device. The video encoding method includes: during video encoding, obtaining an input video sequence; selecting, from candidate video sequence encoding modes, a target video sequence encoding mode corresponding to the input video sequence, the candidate video sequence encoding modes including a constant-resolution encoding mode and a mixed-resolution encoding mode; and encoding each input video frame of the input video sequence according to the target video sequence encoding mode, to obtain encoded data.

Abstract: Aspects of the disclosure provide methods, apparatuses and systems for wireless communication. In some embodiments, a wireless communication system includes processing circuitry. The processing circuitry of the wireless communication system operates based on a first radio access technology. The processing circuitry receives a sidelink control request from a first user equipment that accesses the wireless communication system using the first radio access technology. The sidelink control request requests sidelink control information for a second radio access technology. The processing circuitry then obtains the sidelink control information for the second radio access technology, and provides, via the first radio access technology, the sidelink control information for the second radio access technology, to the first user equipment. The sidelink control information is used to configure a sidelink between the first user equipment and a second user equipment using the second radio access technology.

Abstract: A method and device for sending and receiving virtual goods are provided. A client device may communicate with a server. The client device may receive a receiving link from the server. The server may generate the receiving link in response to an invocation request received by the server from an invoker device that is different from the recipient client device after a total amount of the virtual goods in an invoker account corresponding to the invoker device is successfully transferred to a server account corresponding to the server. The receiving link may include instructions configured to cause the recipient client device to request a virtual goods package from the server. The client device may detect the link is triggered. The client device may request, from the server, in response to link being triggered, the virtual goods package. The client device may accept, by the recipient client device, the virtual goods package by interacting with the server via the receiving link.

Abstract: Provided is a border detection method, server, and storage medium. The method including detecting a plurality of first straight line segments in a to-be-detected image, the to-be-detected image comprising a target region of a to-be-determined border; generating a plurality of first candidate borders of the target region according to the plurality of first straight line segments; obtaining a plurality of second candidate borders of the target region from the plurality of first candidate borders; extracting border features of the plurality of second candidate borders; and obtaining an actual border of the target region from the plurality of second candidate borders according to the border features of the plurality of second candidate borders and a border detection model, the border detection model being used to determine a detected value of each candidate border, and the detected value representing a similarity between each candidate border and the actual border.

Abstract: This application provides a method for constructing a motion information list in video encoding and decoding, an apparatus, and a device, relating to the field of video encoding and decoding technologies. The method includes: obtaining candidate motion information and a length of the motion information list; based on the length of the motion information list, determining target motion information to check for duplicates in the motion information list; checking the motion information list to for duplicates by comparing the target motion information with the motion information list; and updating the motion information list based on whether the motion information list has a duplicate of the target motion information.

Abstract: This application provides an image dehazing method, apparatus, and device, and a computer storage medium. The method includes: in response to obtaining an image dehazing instruction, acquiring a first image and a second image corresponding to a target scene at the same moment. The method also includes calculating, based on a first pixel value of each pixel of the first image and a second pixel value of each pixel of the second image, haze density information of the each pixel; generating an image fusion factor of the each pixel according to the haze density information, the image fusion factor indicating a fusion degree between the first image and the second image; and fusing the first image and the second image according to the image fusion factor to obtain a dehazed image.

Abstract: This disclosure relates to a method for matching a wireless hotspot and a point of interest. The method may include performing semantic normalization preprocessing on a wireless hotspot name and a point of interest name respectively. The method may further include determining a first similarity between the preprocessed wireless hotspot name and the preprocessed point of interest name according to a matching result of consecutive common strings of the preprocessed wireless hotspot name and the preprocessed point of interest name. The method may further include determining a second similarity between the preprocessed wireless hotspot name and the preprocessed point of interest name according to a length difference between lengths of the preprocessed wireless hotspot name and the preprocessed point of interest name. The method may further include determining a matching result of the wireless hotspot and the point of interest according to the first similarity or the second similarity.