429 results about "End system" patented technology

A neural end-to-end system is provided for the face and voice preserving translation of videos. The system is a pipeline of multiple models that produces a video of the original speaker speaking in the target language with modified lip movement to match the target speech, while preserving emphases and prosody of the original speech, and voice characteristics of the original speaker. The pipeline starts with automatic speech recognition including emphasis detection, followed by the translation model. The translated text is then synthesized by a Text-to-Speech model that recreates the original emphases in the target sentence. The resulting synthetic speech is then converted back to the original speakers' voice using a voice conversion model. Finally, to synchronize the lips of the speaker with the translated audio, a generative model generates frames of adapted lip movements which are combined with the audio to produce the final output. The disclosure further describes several use-cases and configurations that apply these techniques to video conferencing, dubbing, low-bandwidth transmission, speech enhancement and assistive technology for the hearing impaired.

The invention relates to the technical field of cloud side end computing, and discloses a cloud side end cooperative computing resource dynamic allocation method based on task-aware migration. The method comprises the following steps: acquiring real-time load characteristics and resource demand characteristics of calculation tasks in a cloud side end system, and dividing task priority queues in combination with task type identifiers; extracting historical execution records of tasks at cloud, edges and terminal nodes, constructing a task execution feature library, and generating a resource demand prediction model in combination with real-time load features; analyzing network transmission time delay characteristics of a cloud end and edge nodes, measuring real-time calculation capability fluctuation data of terminal equipment, and establishing an inter-node resource collaboration degree evaluation matrix; generating an initial migration strategy according to the prediction model and the evaluation matrix, monitoring actual resource occupancy deviation of the task, forming a final decision in combination with a node resource state correction strategy, triggering cross-node migration, and synchronously updating the priority queue and the evaluation matrix.

The invention relates to the technical field of intelligent electric energy meters, in particular to an intelligent electric energy meter system based on an RF net-shaped ad hoc network and a communication method of the intelligent electric energy meter system. The system adopts an intelligent electric energy meter with an integrated gateway function as a gateway node in an RF network and uploads power utilization data and state information to a head end system; independent gateway equipment does not need to be additionally deployed, so that the hardware procurement cost, the installation and debugging cost and the long-term operation and maintenance cost are remarkably reduced; meanwhile, according to the intelligent electric energy meter system communication method based on the RF net-shaped ad hoc network, the optimal main path and the optimal standby path are selected for data transmission by dynamically calculating the comprehensive routing metric value, the reliability of the network is greatly improved through the multi-path redundancy design, and the system has the higher fault self-healing capacity; in addition, the intelligent electric energy meter integrated with the gateway function can monitor and optimize the network topology structure, the data flow distribution and the node resource utilization rate in real time through a built-in network management system, and the data transmission efficiency and the network stability are further improved.

According to the model lightweight deployment method and device based on the hardware performance of the end system, parameters such as the maximum computing power and the batch processing amount are obtained through a static hardware performance evaluation query database, and the actual computing power and the memory are calculated in real time through dynamic operation state monitoring to serve as compression targets; when the hardware resource change exceeds a threshold value, light weight is triggered, and dynamic adjustment of a compression strategy is achieved; in the lightweight process, hardware parameters and model features are fused into a low-dimensional state vector, collaborative compression parameters of a reinforcement learning strategy network output pruning rate, quantization bit width and distillation temperature are input, and finally a lightweight model adaptive to the hardware dynamic environment is generated. The automatic compression process reduces the labor cost, ensures that the lightweight model is accurately matched with the end side hardware constraint, improves the reasoning speed and reduces the energy consumption.

The invention discloses a causal control VLA end-to-end system fusing perception and a world model, and belongs to the technical field of automatic driving. The system comprises a perception and world modeling subsystem; a reasoning and generating subsystem; a verification and collaborative learning subsystem; moreover, the perception and world modeling subsystem is composed of an environment perception module and a world model training module, the reasoning and generation subsystem is composed of a causal modeling module and a diffusion trajectory generation module, and the verification and collaborative learning subsystem is composed of a form verification module and a federal collaborative distillation module. According to the VLA end-to-end system, traditional correlation modeling of a world model is enhanced through causal modeling, so that the world model becomes more accurate; meanwhile, on the basis of traditional verification and updating of the world model, the verification and updating process of the world model generation trajectory is further enhanced through the verification and collaborative learning subsystem, so that the generated trajectory is safer and more reliable.

The invention provides an intelligent customer service system, which comprises an information acquisition module used for identifying a customer identity according to a user login account and obtaining historical data under the account; the grading module is used for establishing a customer portrait model according to the historical data and classifying customer types; the matching module is used for automatically matching a service strategy model version corresponding to the account according to the customer type obtained by the grading module; the knowledge graph response module is used for carrying out deep analysis on client input contents based on the natural language understanding capability of a large language model in combination with a context, a client portrait and historical data after the version of the service strategy model is determined, automatically generating a task list and a semantic tag tree for a dialogue, and continuously optimizing the task list and the semantic tag tree; starting from a problem node of the task list, analyzing a corresponding semantic tag along an optimized and trained path in the knowledge graph, and matching a service processing action; and triggering a back-end system interface associated with the node according to needs, transferring real data, and solving client problems.

The invention discloses an interaction control method and system for an intelligent agent and a front-end and back-end system, and the method achieves the automatic control of the intelligent agent on the front-end and back-end system through the integration of a large language model. The method comprises the following steps: an intelligent agent performs natural language intention recognition through a large language model, automatically selects a proper tool, extracts parameters required by the tool, converts the intention into a target task, generates a structured control instruction, maintains a complete session history and supports multiple rounds of dialogue interaction; the back end receives a user request, calls an intelligent agent to generate a structured instruction, transmits the structured instruction to the front end, and dynamically adjusts a subsequent instruction according to the received front end feedback; and the front end receives the instruction transmitted by the rear end, automatically identifies an instruction format through a regular expression, performs corresponding operation after parameter verification and feeds back a result. According to the method, unified control of the agent on the front-end UI and the back-end business logic is realized, complex multi-round dialogue and multi-step operation are supported, and the method has a wide application prospect.

The invention relates to the technical field of computers, can be applied to the field of science and technology finance / digital medical treatment, and discloses a back-end system creativity transformation method and device, electronic equipment and a storage medium. The method comprises the following steps: deploying a credential back-end system environment by using a credential server and credential middleware, and physically isolating credential traffic and non-credential traffic; carrying out dynamic segmentation scheduling on the traffic of the credential and credential gateway in the credential and credential environment and the traffic of the non-credential and credential gateway in the non-credential and credential environment; deploying a domestic database, and finishing data migration from the original database to the domestic database through an automatic tool chain; segmenting the traffic to a credential and credential gateway in the credential and credential environment for gray verification, and switching a back-end service chain of the credential and credential environment to a domestic database after the verification is passed; and monitoring the operation performance data of the credential environment in real time, and switching the traffic to the non-credential environment and the original database through the global traffic management tool if an exception occurs. According to the method, the risk controllability and smooth transition of creative transformation are realized.

According to the back-end system fault self-recovery method and device, the storage medium and the computer equipment provided by the invention, in the system operation process, multi-dimensional state analysis is performed on the operation data collected by the system micro-service in real time to obtain the operation state; when the operation state is abnormal, the fault analysis model is used for performing fault analysis on the operation data, so that the fault positioning efficiency is improved; after fault data output by the model are obtained, a self-healing strategy corresponding to the fault data is obtained through matching in the dynamic knowledge base, then self-healing operation is executed on the micro-service, and quick response of fault processing is achieved. If the self-healing result is failure, self-healing data in the self-healing operation process is obtained, the self-healing strategy of the micro-service is adjusted to be more adaptive to the actual fault condition, then the self-healing operation is executed on the micro-service again based on the new self-healing strategy until the self-healing result meets the self-healing ending condition, and therefore the flexibility of system fault self-healing is improved; and the influence of the fault on the system stability is reduced to the greatest extent.

The invention discloses a quick data traceability method and system based on multi-source log analysis, and relates to a data processing system or method specially suitable for administrative, commercial, financial, management, supervision or prediction.The method comprises the steps that firstly, the system responds to a service instruction to generate a transaction receipt containing a unique serial number and service attribute parameters; and establishing an association relationship between the two records and generating a to-be-confirmed record and a timestamp. And sending a service message containing the service attribute parameters to a receiving end. When an ownership right reverse search request of a receiving end is received, the system calculates an effective time interval based on current time and a preset time delay threshold value, and screens out a target to-be-confirmed record which is matched with a service attribute parameter and generates a timestamp in the interval. And finally, extracting the unique serial number from the target record, generating a confirmation receipt and sending the confirmation receipt to a receiving end system. The method is used for reducing the calculation overhead and false alarm rate of accurately positioning specific abnormal transactions in massive logs by a system, and improving the confidence and timeliness of an automatic traceability result.

The invention discloses a foreground and background state information intelligent synchronization method in project management. The method comprises the following steps: S1, constructing a three-layer architecture of a foreground system, a state storage center and a background system; s2, initializing a state; s3, realizing state synchronization of the foreground system and the background system through at least one synchronization mode; s4, constructing a version control and conflict resolution mechanism, when the state of each project is updated, the foreground system and the background system carry version numbers, checking the consistency between the client version number of the foreground system and the server version number of the background system when the update is submitted, if the client version number of the foreground system and the server version number of the background system are consistent, executing state updating and progressively increasing the version number, and if the version number of each project is not consistent, executing state updating and progressively increasing the version number; if yes, judging that the data conflict exists, automatically executing a preset conflict solving strategy, and broadcasting a resolution result to a foreground system and a background system. The method has the advantages that the item switching state is not lost, background state change and foreground real-time response are achieved, and conflicts are automatically eliminated without manual intervention.

The invention relates to the technical field of online control, in particular to a multi-parameter collaborative optimization control method based on a large-model refrigeration station end system, and the method specifically comprises the steps: evaluating the working coupling degree of a refrigeration station; the working performance of the refrigeration station end system is deeply evaluated, and the optimization direction of the control strategy is pre-estimated; calculating load weights of new and old tail ends, actively compensating hydraulic unbalance between the new and old tail end systems, and selecting a combined optimization control mode of the tail end system according to an energy efficiency difference coefficient; and continuously monitoring the compatibility of the new and old systems at the tail end of the refrigeration station, and outputting differential control parameters of the system at the tail end of the refrigeration station when the monitoring result is incompatible. According to the method, the problem that in the prior art, a dynamic collaborative optimization mechanism for new and old terminals of the refrigeration station aiming at multi-dimensional parameter differences is lacked is solved.

The invention discloses a 350MW thermal power generating unit cold end system optimization method based on big data, and relates to the field of electric power. The method comprises the following steps: S1, data collection: selecting historical data during the operation period of a unit as modeling data; s2, data preprocessing: preprocessing historical data; s3, constructing a model: constructing a BP neural network model based on the preprocessed data, and optimizing a weight and a threshold value of the BP model through a genetic algorithm and a particle swarm algorithm; s4, calculating the optimal vacuum degree: calculating the optimal vacuum degree under different working conditions according to the optimized BP neural network model; and S5, cold end system control: sending the optimal vacuum degree and a circulating water pump frequency conversion instruction to a DCS (Distributed Control System) through an MODBUS communication protocol to realize optimal control of the cold end system. The method has the beneficial effects that the cold end system of the 350MW thermal power generating unit is optimized by adopting big data analysis and an intelligent algorithm, so that the comprehensive effects of improving the operation efficiency, reducing the energy consumption, saving the cost and reducing the emission are realized.

The invention discloses an interactive terminal system for artificial intelligence recruitment, and relates to the technical field of artificial intelligence and human resources. The system comprises an intelligent interaction module, a demand analysis module, a data storage module, an intelligent matching engine, a post issuing module, an intelligent evaluation module, a process management module and a user feedback and model optimization module. The system receives an unstructured recruitment demand through a multi-modal interface, analyzes and generates structured post data based on a natural language processing technology, realizes accurate man-post matching through a multi-dimensional matching algorithm, and automatically completes post release, skill evaluation, video interview analysis and process management. According to the method, the problems of inaccurate demand understanding, low matching efficiency and tedious process in traditional recruitment are effectively solved, intelligent and automatic management of the whole recruitment process is realized, and the recruitment efficiency and talent matching accuracy are remarkably improved.

Devices, systems, and process for generating fixture specific models for use in adjusting pricing of real-time betting lines are described. A system may include a front-end system including an event-activity-fixture (EAF) server that adjusts betting lines based on results of simulations generated by an EAF simulation server (EAFSS). The EAFSS generates the simulations using fixture specific models that have been generated based upon adaptations of generic models, where the adaptations occurring using historic and real-time EAF data and fixture specific modeling data. The fixture specific models adapted from one or more generic models are leveled and stored in a database for use by the EAFSS on a real-time basis as the EAF occurs. A server for adapting the generic models instantiate one or more computer engines including a model adaptation engine, a generic modeling engine, an EAF data engine, and an FSM leveling engine.

The invention provides an information fusion method and device for a vehicle-road cooperation end-to-end system, and relates to the technical field of automatic driving, and the method comprises the steps: extracting a second target level feature, a second lane level feature, a second scene level feature and a second track level feature through the vehicle-end end-to-end system; after space-time alignment is carried out on the first target level feature and the second target level feature, the first lane level feature and the second lane level feature, the first scene level feature and the second scene level feature, and the first track level feature and the second track level feature, cross-view track feature fusion is carried out; and finally, path planning is carried out based on all the finally obtained fusion features and the second aerial view features. According to the method, space-time alignment and fusion of the target level features are considered, and space-time alignment and fusion of the lane level features, the scene level features and the track level features are also considered, so that the accuracy of final path planning is improved.

Instead of incurring interface errors caused by power drops, continuously track all the channels between the controller and the storage device (both in-band and side-band channels). The tracking will include extracting relevant indications and use a prediction model that correlates between the tracked indications to later occurrence of power drop events. In response to the prediction results, the system may perform different rehabilitation operations (countermeasures). The controller will monitor different indications from the storage element and power supply to predict a signal integrity degradation marginality event on the interface. The monitoring and mitigation strategy relies on an open-ended system. The controller does not employ a real-time continuous return channel from the storage device that may signal pass / fail conditions. Given the nature of an open-ended system and tolerance of calibration for the data gathering, processing and inference system, the controller will have to cope also with false positive events.

The invention belongs to the field of vehicle networking communication, and particularly relates to a hybrid network communication system for network connection cooperative driving, which comprises the following steps: analyzing urban road straightaway and highway straightaway scenes including scene composition and scene traffic flow; deploying RSU and MEC calculation units according to scene analysis; establishing a multi-mode communication and hybrid network topology structure; a hybrid network cooperative lane changing communication system is established according to multimode communication and a hybrid network topology structure, and comprises a vehicle end system, a road side system and a cloud control platform; based on a hybrid network cooperative lane changing communication system, executing a CAV cooperative lane changing driving service; the continuity of cross-network communication is realized, and an efficient, low-delay and high-reliability cooperative driving task is realized.

The invention discloses a front-end event processing method and related equipment. The method comprises the following steps: acquiring operation binary data sent by a front-end main thread; analyzing the operation binary data, and classifying a plurality of operation messages obtained by analysis into an established message queue; the operation messages in the message queue are adjusted in real time according to a preset message optimization processing rule, and event cycle polling is started to process the message queue; and sending rendering data generated by polling processing to the front-end main thread to realize front-end rendering. According to the invention, efficient and stable front-end event processing can be realized, and the performance and response speed of a front-end system can be comprehensively improved.

The invention discloses a new energy tracking / networking converter capacity configuration method based on a power angle stability boundary, and the method comprises the steps: firstly, building a new energy sending end system simulation model containing a networking converter, and calculating a steady-state operation point of a system; then, carrying out dynamic equivalence on the network construction type converter and a synchronous generator to obtain equivalent machine parameters, and simplifying an original simulation system into an equivalent machine and a network following type new energy sending end system; deriving mechanical power and electromagnetic power of the equivalent machine before, during and after the fault to obtain a power angle characteristic curve of the equivalent machine; calculating the acceleration area and the maximum deceleration area of an equivalent machine under different proportions of the network construction type converter, and obtaining the proportion of the network construction type new energy under the stable critical power angle of the system by adopting a dichotomy method; simulation is carried out based on a PSASP platform, and the rationality of the proposed configuration principle is verified. According to the scheme provided by the invention, the network construction investment of the system can be reduced while the safety and stability of the system are ensured, and the admitting ability of the power grid to new energy is improved.

The invention discloses a digital operation management platform oriented to enterprise transformation and upgrading and a construction method thereof, and belongs to the field of enterprise digital transformation. The platform comprises a front-end system cluster which comprises a safety management module, a production management module, an asset management module, a material management module, a collaborative office module, a production real-time module, a public foundation module, a system management module and a data governance module. The construction method comprises the following steps: S1, constructing a unified data interface architecture in a data governance server, accessing a multi-source heterogeneous data source and caching an original data packet; s2, cleaning and converting data based on the unique field identifier, and generating a structured data table; s3, transmitting the data to a data fusion center node, performing partition storage, deconstructing semantics, and generating an index metadata set; s4, constructing a multi-path service route in the data service distribution server, and generating a static index table and a dynamic index flow; and S5, pushing the data to a front-end system cluster through an interface service layer, and calling the data by a user terminal to realize index-driven data view construction. The method has the beneficial effects that the data integration efficiency and the operation intelligence level are improved.

The invention relates to the technical field of interaction based on a graphical user interface, in particular to an auxiliary learning method and system for old people. The method comprises the following steps: deploying a front-end monitoring service in a user intelligent machine; deploying a simulation mirror image system in a back-end system; when intervention is carried out, the operation purpose of the elderly user is obtained, the simulation mirror image system retrieves a target operation path corresponding to the operation purpose from a pre-established path library, and the target operation path is implemented to carry out path verification: in the process of executing the simulation touch event, a jittering simulation touch event is added, and when the simulation touch event is executed, the jittering simulation touch event is executed; calling an addressing operator based on a large model to carry out addressing on the operation path so as to obtain a final operation path which considers a jitter factor and meets a final state; and feeding back the final operation path to the elderly user, so that the elderly user can learn on the intelligent machine according to the steps. The method can more accurately assist the elderly user in learning to use the smart phone.

The invention provides a data synchronization method and system, electronic equipment, a storage medium and a program product. The method is applied to an agent layer, and comprises the following steps: acquiring a change event of user data of a source end system, and generating an event stream according to the change event; screening out target user data with changed attribute values based on timestamp information in the event stream; synchronizing the target user data to a target end system; wherein the agent layer is decoupled from the source end system, and the agent layer is decoupled from the target end system. The method has the advantages that thorough decoupling of the agent layer and the source end system / target end is achieved through agent layer zero intrusion event capturing and incremental timestamp screening, meanwhile, a traditional full-amount scheme is replaced with an incremental synchronization mechanism with millisecond-level delay and thousand-level throughput, resource consumption is remarkably reduced, and the method is adaptive to enterprise-level high-availability scenes.

A sea area stratum shield tunneling grouting intelligent management and control system and method comprises a storage subsystem, a pipeline subsystem and a grouting front-end subsystem, and a cement mortar storage system and an accelerator storage system are connected with a synchronous grouting front-end system and a secondary grouting front-end system through a synchronous grouting pipeline and a secondary grouting pipeline. And the synchronous grouting front-end system and the secondary grouting front-end system control the flow and the speed in the pipeline through an electric control valve. Automatic control and synchronous grouting evaluation indexes are adopted, the effects of synchronous grouting and secondary grouting can be effectively improved, the grouting quality is improved, building gaps can be filled in time, stratum deformation can be effectively controlled, the settling volume of a tunnel is reduced, and the anti-seepage and waterproof effects of the tunnel can be improved; the synchronous grouting accelerator is automatically added, the adding rate can be accurately controlled, and accurate metering can be achieved; the residual amount of the grouting materials is automatically checked after each ring in the shield tunnel is tunneled, and the ground terminal can timely prepare the quantity of the materials of the next ring according to real-time data.

An AI decision based multiple telecommunication endpoints system is provided, including a telecommunication endpoint device and an artificial intelligence decision controller. The telecommunication endpoint device performs a communication test with the radio frequency communication device under test. The artificial intelligence decision controller is electrically connected to the telecommunication terminal device to control the communication test, and performs an efficiency analysis on the result of the communication test, and generates a decision instruction according to the result of the efficiency analysis. The artificial intelligence decision controller can centrally control, replace and analyze multiple artificial intelligence modules and generate a large number of telecommunication endpoints device signal sending and receiving behaviors, providing a telecommunication endpoints system with self-adaptive adjustment process, parameters, result analysis and reasonable cost as a test for the development phase, deployment phase, or maintenance phase of the telecommunications equipment product development process.

The present invention provides a database access method and related devices based on an edge-cloud-end system. The method includes constructing an edge-cloud-end system of a terminal, an edge gateway, and a cloud server; deploying an enterprise metadata knowledge base, an AI model, and an enterprise database in the edge-cloud-end system based on customer requirements; receiving a demand prompt word input by a user through an intelligent agent in the terminal; obtaining a routing policy through the intelligent agent based on the deployment scenarios of the enterprise metadata knowledge base, the AI model, and the enterprise database and with the principle of giving priority to local gateway processing; and programmatically processing the demand prompt word based on the routing processing policy, the edge gateway, and the cloud server. The database access method of the present invention borrows an AI model for code generation, reduces the development difficulty, and the enterprise metadata knowledge base, the AI model, and the enterprise database are flexibly deployed based on customer requirements, and the terminal intelligent agent can flexibly route. The present invention has better adaptability and can better meet the enterprise requirements.

The application discloses a kind of causal control VLA end-to-end systems of fusion perception and world model, belong to the technical field of automatic driving;The system includes: perception and world modeling subsystem;Reasoning and generation subsystem;Verification and collaborative learning subsystem;And, perception and world modeling subsystem is constituted by environment perception module and world model training module, reasoning and generation subsystem is constituted by causal modeling module and diffusion trajectory generation module, verification and collaborative learning subsystem is constituted by formal verification module and federal collaborative distillation module;The VLA end-to-end system of the application, through causal modeling, enhances the traditional correlation modeling of world model, so that the world model becomes more accurate;At the same time, on the basis of traditional verification and updating of world model, the verification and updating process of trajectory generated by world model is further strengthened through verification and collaborative learning subsystem, so that the generated trajectory is more safe and reliable.

The invention discloses a front-end system reconfiguration version control method and device, equipment, a storage medium and a product, and relates to the technical field of system reconfiguration, and the method comprises the steps: receiving a first access request of a user for a target menu in a parallel operation stage of an old system and a new system, and obtaining the identity information of the user; judging whether an access system corresponding to the first access request is the new system or not based on the identity information and a version mapping table; when it is judged that the access system is the new system, a micro-front-end frame is called to load sub-application static resources, corresponding to the target menu, in the new system to a browser for rendering, and a special area page is obtained, and forwarding an interface calling request triggered by the user in the special area page to a new back-end service through a new system gateway. According to the invention, the technical risk and integration complexity of large-scale front-end system reconstruction can be reduced.

The invention discloses a front-end system performance automatic monitoring analysis processing method and a terminal, and belongs to the technical field of Internet, and the method comprises the steps: monitoring a front-end system in real time according to preset performance index parameters, and capturing corresponding index data; performing integration, data format unified processing and data cleaning processing on the captured index data; according to a preset analysis demand, generating a corresponding visualized billboard diagram from each piece of processed index data; carrying out statistics and analysis on response time, resource loading time and user behavior conditions of each index data of the generated visual billboard graph, and generating an analysis result; and performing performance optimization on the index data which does not conform to a preset value in the generated analysis result, and outputting an optimization result. According to the method, the front-end system can be automatically monitored, counted and analyzed, and is displayed through a visual interface, so that problem positioning and targeted optimization are conveniently and quickly carried out, the improvement of system performance is facilitated, and convenience is provided for the use of a user.

An end-to-end system and method for detecting objects of interest in geospatial imagery where the initial query comprises only an abstract of the object. A self-supervised student-teacher platform enables training an accurate model with a dataset of examples of classes of objects that may occur within geospatial imagery. Algorithms are applied to automate searching patterns in the dataset that actualize the abstract to cause the model to return images responsive to the abstract. The result may be iteratively refined so that images ultimately yielded by the search are examples of the object of interest.