Method for preventing false triggering of live streaming risk control, apparatus, device and product

Abstract

This application relates to a method, apparatus, device, and product for preventing accidental triggering of live streaming risk control. The method includes: responding to a virtual live streaming start command and starting an e-commerce live streaming room on an e-commerce platform for executing virtual live streaming activities; obtaining a script list from a script server through a script generation service, the script list containing script texts corresponding to different business stages in the same live streaming business process; generating corresponding human-image voiceover videos based on template videos using a video generation service, the template videos reaching a preset regulated duration and each image frame containing facial images captured from the same person; and pushing each human-image voiceover video to the e-commerce live streaming room according to the live streaming business process through a virtual camera driving service. This application can reduce the frequency of accidental triggering of interference behavior by the risk control system against virtual live streaming activities in the e-commerce live streaming room, and improve the stability and security of virtual live streaming activities.

Description

Specification 1 Live Streaming Risk Control and Accidental Touch Prevention Method, Device, Equipment, and Product Technical Field This application relates to the field of e-commerce information security technology, and particularly to a live streaming risk control and accidental touch prevention method, device, equipment, and product. Background Technology E-commerce platforms all deploy risk control systems to identify various non-compliant behaviors of e-commerce users, so as to take corresponding regulatory actions in a timely manner to maintain platform information security. With the development of online live streaming technology, e-commerce platforms and online live streaming have been deeply integrated. Various e-commerce platforms are also gradually strengthening security control for e-commerce live streaming activities, such as identifying the behavior of merchants using robots to mechanically repeat broadcasts, so as to promptly detect illegal broadcasting behavior and ensure the quality of live streaming content generated by e-commerce live streaming. Virtual human live streaming, also known as digital human live streaming, is increasingly widely used in online live streaming. Unlike robot broadcasting, virtual human live streaming uses text or voice to drive the generation of video streams for live streaming. This video stream contains image content corresponding to the information content verbally delivered by a specific person. The information content to be verbally delivered can be pre-customized or determined in real time. Therefore, virtual human live streaming is a relatively real-time video dynamic generation technology. When the video stream corresponding to the virtual human is pushed to the live streaming room for playback, a virtual live streaming activity can be executed. The video stream in a virtual live streaming activity is usually generated based on a certain length of source video. However, the number of image frames in the source video is always limited. When generating the video stream required for the live stream for a long time, due to the technological lag of the risk control system, the system may easily misidentify the video stream in the virtual live streaming activity as robot-assisted broadcasting, and make incorrect interventions, such as stopping or banning the broadcast. This can lead to the inability to conduct e-commerce live streaming activities normally, causing huge economic losses to e-commerce stores and platforms, damaging the user experience, and limiting the development of virtual human live streaming as a new technology. Therefore, it is necessary to improve the related technologies of virtual live streaming activities by addressing the technical deficiencies of the risk control system. The purpose of this application is to provide a method, device, equipment, and product for preventing accidental touches in live streaming risk control. According to one aspect of this application, a method for preventing accidental touches in live streaming risk control is provided, comprising: responding to a virtual live streaming start command and starting an e-commerce live streaming room on an e-commerce platform for executing virtual live streaming activities; obtaining a script list from a script server through a script generation service, wherein the script list contains script texts corresponding to different business stages in the same live streaming business process; and, through a video generation service, calling a video server to generate a human-image voiceover video corresponding to each script text based on a template video, wherein the template video reaches a preset regulated duration and each of its image frames contains a facial image captured based on the same person; and using a virtual camera driver service.According to the live streaming business process, the corresponding human-image spoken videos of each script text in the script list are pushed to the e-commerce live streaming room to implement the virtual live streaming activity. According to another aspect of this application, a live streaming risk control and accidental touch prevention device is provided, comprising: a live streaming response module, configured to respond to a virtual live streaming start command and start an e-commerce live streaming room in an e-commerce platform for executing virtual live streaming activities; a script acquisition module, configured to acquire a script list from a script server through a script generation service, the script list containing script texts corresponding to different business stages in the same live streaming business process; a video acquisition module, configured to call a video server through a video generation service to generate human-image spoken videos corresponding to each script text based on a template video, the template video reaching a preset regulated duration, and each image frame containing a facial image captured based on the same person; and a live streaming push module, configured to push the corresponding human-image spoken videos of each script text in the script list to the e-commerce live streaming room to implement the virtual live streaming activity through a virtual camera driver service according to the live streaming business process. According to another aspect of this application, a computer device is provided, including a central processing unit and a memory. The central processing unit is used to call and run a computer program stored in the memory to execute the steps of the live streaming risk control and accidental touch prevention method described in this application. According to another aspect of this application, a non-volatile readable storage medium is provided, which stores a computer program implemented according to the live streaming risk control and accidental touch prevention method in the form of computer-readable instructions. When the computer program is called and run by a computer, it executes the steps included in the method. According to another aspect of this application, a computer program product is provided, including a computer program / instructions. When the computer program / instructions are run by a processor, it executes the steps of the live streaming risk control and accidental touch prevention method. Compared with the prior art, when a live streaming user starts a live streaming room to execute a virtual live streaming activity, this application first calls a script list from a script server through a script generation service to obtain the script texts corresponding to each business link of the same live streaming business process. Then, a video generation service drives a video server to instantly generate human-image voice-over videos corresponding to each script text based on template videos that have reached a preset regulatory duration. Finally, a virtual camera driving service pushes these human-image voice-over videos to the e-commerce live streaming room to implement the virtual live streaming activity. This demonstrates the advantages of this application in several aspects, including: In terms of the dynamic recognition mechanism, template videos conforming to regulated durations can enrich and generalize the motion features of facial images in the corresponding spoken video scripts, reducing the frequency of repeated motion features detected by the risk control system; in terms of the static recognition mechanism, the virtual camera-driven service avoids the inherent flaw of the risk control system's mechanical judgment of robot broadcasting behavior based on data sources. Therefore, through the combination of these two aspects,This application enables virtual live streaming activities to effectively avoid being misidentified as robot broadcasting by e-commerce platform risk control systems, both dynamically and statically. This reduces the frequency of false interventions by risk control systems targeting virtual live streaming activities in the e-commerce live streaming room, improves the stability and security of virtual live streaming activities, avoids unnecessary economic losses for the e-commerce stores hosting the live streaming room, and also safeguards the application of virtual human live streaming in the e-commerce field, removing obstacles to its application. Brief Description of the Drawings: To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Figure 1 is an exemplary network architecture of an e-commerce platform in this application; Figure 2 is an exemplary network architecture supporting a digital broadcast control system in this application; Figure 3 is a flowchart illustrating the live streaming risk control and accidental touch prevention method in an embodiment of this application; Figure 4 is a flowchart illustrating the insertion of human-image spoken video in response to dynamic insertion instructions in an embodiment of this application; Figure 5 is a flowchart illustrating the determination of inserted text based on user input information in an embodiment of this application; Figure 6 is a flowchart illustrating the creation of template video in an embodiment of this application; Figure 7 is a flowchart illustrating the creation of template video guided by the interference rate of the risk control system in an embodiment of this application; Figure 8 is a flowchart illustrating the generation of corresponding human-image spoken video by a video server based on spoken text in an embodiment of this application; Figure 9 is a structural diagram of the live streaming risk control and accidental touch prevention device in an embodiment of this application; Figure 10 is a structural diagram of the computer equipment in an embodiment of this application. In the network architecture shown in Figure 1, the e-commerce platform 82 is deployed on the Internet to provide corresponding services to its users. Similarly, the devices 80 of the merchant users and the devices 81 of the consumer users of the e-commerce platform 82 are also connected to the Internet to use the services provided by the e-commerce platform. For example, an e-commerce platform can configure a live streaming system to provide live streaming services to merchants in its online stores. In this live streaming service, the merchants are the broadcasters. Broadcasters can launch live streaming activities, including live streams with real people or virtual live streams based on virtual or digital humans, by starting their own live streaming room. Consumers can also act as viewers, entering the live streaming room to receive video and information streams and interact with the broadcasters to participate in the live streaming activities. An exemplary e-commerce platform 82, leveraging internet infrastructure, provides supply and demand matching for products and / or services to the general public. In e-commerce platform 82…Products and / or services are provided as commodity information. For simplicity, the concepts of "commodity" and "product" are used in this application to refer to products and / or services on e-commerce platform 82. Specifically, these can be physical products, digital products, tickets, service subscriptions, and other offline services. Various entities in reality can access e-commerce platform 82 as users and use the various online services provided by e-commerce platform 82 to achieve the purpose of participating in the business activities facilitated by e-commerce platform 82. These entities can be natural persons, legal persons, or social organizations. Corresponding to the two types of entities in business activities—merchants and consumers—e-commerce platform 82 has two corresponding categories of users: merchant users and consumer users. Entities in the product circulation chain of business activities, including manufacturers, sellers, retailers, and logistics providers, can all use online services on e-commerce platform 82 as merchant users. Consumers in business activities, including actual or potential consumers, can use online services on e-commerce platform 82 as consumer users. In actual business activities, the same entity can act as both a merchant user and a consumer user; this should be interpreted flexibly. The infrastructure used to deploy the e-commerce platform 82 mainly includes a backend architecture and frontend devices. The backend architecture runs various online services through a service cluster, including middleware or frontend services for the platform, services for consumers, and services for merchants, to enrich and improve its service functions. The frontend devices mainly cover the terminal devices used by users as clients to access the e-commerce platform 82, including but not limited to various mobile terminals, personal computers, and point-of-sale devices. For example, merchant users can use their terminal device 80 to enter product information for their online stores or use the interfaces opened by the e-commerce platform to generate their product information; consumer users can use their terminal device 81 to access the webpage of the online store implemented by the e-commerce platform 82, trigger the shopping process by using the shopping button provided on the webpage, and call various online services provided by the e-commerce platform 82 during the shopping process to achieve the purpose of placing an order. In some embodiments, the e-commerce platform 82 can be implemented by a processing facility including a processor and memory, which stores a set of instructions that, when executed, cause the e-commerce platform 82 to perform the e-commerce and support functions involved in this application. Processing facilities can be servers, clients, network infrastructure, mobile computing platforms, cloud computing platforms, fixed computing platforms, or other computing platforms.It also provides electronic components, merchant devices, payment gateways, application developers, marketing channels, transportation providers, customer devices, and point-of-sale devices for the e-commerce platform 82. The e-commerce platform 82 can provide online services such as cloud computing services, Software as a Service (SaaS), Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Desktop as a Service (DaaS), Hosted Software as a Service, Mobile Backend as a Service (MBaaS), and Information Technology Management as a Service (ITMaaS). In some embodiments, the various functional components of the e-commerce platform 82 can be implemented to operate on various platforms and operating systems. For example, for an online store, its administrator user enjoys the same or similar functions regardless of whether it is on iOS, Android, HomonyOS, or a web page. The e-commerce platform 82 can implement corresponding independent websites for each merchant to run their respective online stores, providing merchants with corresponding business management engine instances for them to establish, maintain, and run one or more online stores on one or more independent websites. A business management engine instance can be used for content management, task automation, and data management of one or more online stores. It can be configured through interfaces or built-in components to support various specific business processes within the online store, enabling business activities. Independent websites are the infrastructure of the 82 e-commerce platform, which offers cross-border services. Merchants can maintain their online stores more centrally and independently based on independent websites. Independent websites typically have dedicated domain names and storage space, and different independent websites are relatively independent. The 82 e-commerce platform can provide standardized or customized technical support for a large number of independent websites, allowing merchants to customize a business management engine instance that suits their needs and use it to maintain one or more of their online stores. Online stores can be configured and maintained in the backend by merchant users logging into their Business Management Engine instance as administrators. Supported by the various online services provided by the e-commerce platform 82's infrastructure, merchant users can configure various functions within their online stores and view various data as administrators. For example, merchant users can manage various aspects of their online stores, such as viewing recent online store activities, updating the online store product catalog, managing orders, recent visit activity, and total order activity. Merchant users can also view more detailed information about their business and visitors to their online store by obtaining reports or metrics, such as displaying a sales summary of the merchant's overall business, specific sales and engagement data from promotional sales and marketing channels, etc. The e-commerce platform 82 can provide communication facilities and associated merchant interfaces for providing electronic communications and marketing.For example, electronic messaging aggregation facilities can be used to collect and analyze communication interactions between merchants, consumers, merchant devices, customer devices, point-of-sale devices, etc., aggregating and analyzing communications, for example, to increase the potential for product sales. For instance, a consumer may have product-related questions, which could lead to a dialogue between the consumer and the merchant (or an agent based on an automated processor representing the merchant), where the communication facility is responsible for the interaction and providing the merchant with analysis on how to increase the probability of sales. In some embodiments, applications suitable for installation on terminal devices can be provided to serve the access needs of different users, enabling various users to access e-commerce platform 82, such as the merchant backend module of online stores in e-commerce platform 82, by running the application on their terminal devices. During the process of realizing business activities through these functions, e-commerce platform 82 can implement various functions related to realizing business activities as middleware or online services and expose corresponding interfaces. Then, toolkits corresponding to the interface access functions are embedded into the application to achieve functional expansion and task implementation. The business management engine can include a series of basic functions and expose these functions to online services and / or application calls via APIs. Online services and applications use the corresponding functions by remotely calling the corresponding APIs. With the support of various components of the Business Management Engine instance, the e-commerce platform 82 can provide online shopping functionality, enabling merchants to connect with customers in a flexible and transparent manner. Consumers can select items online, create product orders, provide the delivery address in the order, and complete payment confirmation. Merchants can then review and complete or cancel orders. The review component included with the Business Management Engine instance ensures compliant use of business processes, guaranteeing that orders are suitable for fulfillment before actual execution. Orders may sometimes be fraudulent and require verification (e.g., ID checks). A payment method requiring merchants to wait to ensure receipt of funds can mitigate this risk, and so on. Order risks may arise from fraud detection tools submitted by third parties through order risk APIs, etc. Before fulfillment, merchants may need to obtain or wait to receive payment information to mark the order as paid before preparing to deliver the product. Such situations can all be reviewed accordingly. The review process can be implemented by the fulfillment component. Merchants can leverage fulfillment components to review and adjust operations, and trigger related fulfillment services. These include: manual fulfillment services, used when merchants select and pack products into boxes, purchase shipping labels and enter tracking numbers, or simply mark items as fulfilled; custom fulfillment services, which can define email notifications; API fulfillment services, which can trigger third-party applications to create fulfillment records; and legacy fulfillment services.It can trigger custom API calls from the business management engine to third parties; gift card fulfillment services. It can generate and activate gift cards. Merchants can use an order printer application to print shipping documents. The fulfillment process can be executed when items are packaged in boxes and ready for shipment, tracking, delivery, and consumer verification. The e-commerce platform can also deploy a risk control system to perform security detection on the network access behavior of merchant users and / or consumer users during e-commerce or live streaming activities, promptly identify non-compliant operations, and implement corresponding technical intervention measures to ensure the healthy operation of the e-commerce platform. The computer devices of the merchant users (i.e., the broadcasters) in e-commerce live streaming can run computer program products implemented according to the live streaming risk control and accidental touch prevention method of this application, serving as a digital broadcast control system, to avoid the risk control system misidentifying the operations performed by merchant users during live streaming activities as non-compliant operations, and reduce the frequency of the risk control system interfering with live streaming activities. To support merchant users, i.e., livestream users, in running the computer program products of this application, a digital broadcast control system network architecture as shown in Figure 2 can be provided. This network architecture includes a script server, a video server, a livestream server in the e-commerce platform 82 for supporting livestreaming, and the terminal devices used by the livestream users, i.e., livestreaming devices. The script server can provide relevant text, such as script text and interstitial text, to the livestream devices of the livestream users; the video server can help the livestream devices of the livestream users generate corresponding human-image voiceover videos; the livestream server of the e-commerce platform can send the videos pushed by the livestream devices of the livestream users, such as human-image voiceover videos, to the e-commerce livestream room. This allows viewers in the e-commerce livestream room to receive and play the corresponding human-image voiceover videos, enabling the livestream users to achieve their goal of conducting livestreaming activities through their livestream devices. When livestreaming through their e-commerce livestream room, livestream users can conduct livestreaming activities with real people or virtual livestreaming activities based on virtual humans, i.e., digital humans. During virtual livestreaming activities, through multi-party interaction with the script server, video server, and livestream server, they can push specified scripts and human-image voiceover videos corresponding to specified human images to the e-commerce livestream room. HK 20115172 B Specification 7 Please refer to Figure 3. In some embodiments, the live streaming risk control and accidental touch prevention method of this application can be implemented as a computer program product, running on the terminal device of the broadcaster user, to construct a digital broadcast control system, helping the broadcaster user to implement playback control of virtual live streaming activities in the e-commerce live streaming room of the e-commerce platform. The method includes: Step S5100, responding to the virtual live streaming start command, starting the e-commerce live streaming room on the e-commerce platform used to execute virtual live streaming activities; After the broadcaster user runs the computer program product of this application on their terminal device, they can enter the e-commerce live streaming room they registered on the e-commerce platform through a preset control method.In order to start the live broadcast. For example, a broadcaster can click the "Start Live Broadcast" button on the management page of their online store registered on an e-commerce platform, thereby triggering a virtual live broadcast start command. The background process of the computer program product of this application responds to the virtual live broadcast start command and, according to its default business logic, interacts with the live broadcast server of the e-commerce platform to start the broadcaster's e-commerce live broadcast room, so as to execute the virtual live broadcast activity through the e-commerce live broadcast room. Step S5200: Obtain a script list from the script server through a script generation service. The script list contains script texts corresponding to different business stages in the same live broadcast business process. To achieve efficient operation, the computer program product of this application runs a script generation service and communicates with the script server through this script generation service. Therefore, before starting to push the virtual live broadcast activity, the script generation service can first obtain the script list corresponding to the implementation of the virtual live broadcast activity from the script server. The script server has the capability to generate text corresponding to the scripts required for virtual live streaming activities for broadcasters. It can provide corresponding texts to the script generation service through various implementation methods, pre-storing various texts in a corresponding database. When the script generation service needs to call upon these texts, it retrieves the matching text from the database and returns it to the script generation service. For example, in one embodiment, the script server can maintain a live streaming business process database. This database stores script texts corresponding to different business stages within various live streaming business processes. Broadcasters can retrieve a specific list of scripts corresponding to a particular live streaming business process from the script generation service. The script server simply needs to retrieve the script texts corresponding to each business stage in the corresponding live streaming business process from the database to construct the script list and push it to the script generation service. The live streaming business process consists of multiple business stages. Each business stage's corresponding scripts implement the output of the expected information type for that stage, achieving phased information dissemination. The information disseminated at each stage constitutes the overall information disseminated throughout the entire live streaming business process, achieving the intended dissemination purpose. The communication role of each business step in the live streaming process can be flexibly customized. For example, typical live streaming scripts related to recommending products from online stores include: the opening introduction, the pain point explanation, the product selling point presentation, the order closing, and the attention attraction stage. Each business step can pursue its expected communication effect through corresponding scripts, and the scripts used for different business steps are generally different. HK 20115172 B Specification 8 Multiple script texts can be set for each business step. When constructing the script list for a live streaming business process, the script server randomly selects one script text from the multiple script texts corresponding to each business step in the live streaming business process and adds it to the script list. In another embodiment,The script server can also maintain an advertising script database. After the broadcaster's terminal device submits product information to the script server, the script server matches one or more corresponding promotional script texts from the advertising script database based on the product information, forming a script list, which is then returned to the broadcaster. This advertising-related script list can be activated by the broadcaster inputting the product to be advertised. Once the broadcaster confirms the product, they submit it to the script server either through a script generation service or directly. The script server retrieves the product information, including images and / or text, from the e-commerce platform's product database, and then matches the corresponding script texts from the advertising script database to form the script list. Similarly, in another embodiment, the script server can also maintain a customer service Q&A database. When a viewer provides input text related to a question in the e-commerce live stream, the script server matches one or more corresponding answer script texts from the customer service Q&A database based on the input text, forming a script list, which is then returned to the broadcaster. Similarly, this script list can be obtained by the script generation service after detecting the chat history in the e-commerce live stream to obtain the input text of the audience users, and then interacting with the script server. Alternatively, the live stream server can interact with the script server to determine the input text corresponding to the user's question, and then the live stream server retrieves it from the script server and pushes it to the broadcaster user to drive the generation of the corresponding human-image broadcast video. In another embodiment, for the various script texts required by the broadcaster user, the script server can use a well-trained neural network model, such as various large language models, to generate the corresponding script list based on the basic material text and prompt text submitted by the broadcaster user to the script server, and directly return it to the script server. This method can provide not only the script list required for the live stream business process, but also script lists corresponding to advertisements and customer service Q&A. The script list can contain single script texts or multiple script texts, depending on the results generated by the neural network model and specific business needs. In embodiments that flexibly adapt the above various embodiments, the script list obtained by the terminal device from the script server through the script generation service can not only contain a pre-defined script list corresponding to the same live broadcast business process, which only includes the script text corresponding to each business link in the live broadcast business process, but also embed advertising script text into the script list corresponding to the same live broadcast business process, and treat the latter as a business link in the live broadcast business process, i.e., an advertising insertion link. In other words, the live broadcast business process includes an advertising insertion link, and therefore the script list also includes the script text corresponding to the advertising business link. This can promote a more natural and smooth advertising insertion effect when generated subsequently based on the script list. In further enriched embodiments,The script server can respond to the script generation service's request to obtain the script list. First, it determines the script list corresponding to the live broadcast business process specified in the request, as well as the product information of the product to be advertised. Then, it uses each script text in the script list as a reference text, calls a preset prompt text template, and fits it into the product information to become the model prompt text. It inputs the reference text and the model prompt text into the large language model, controls the large language model to associate the reference text to generate the script text of the product to be advertised, and determines the position of the script text in the script list of the live broadcast business process. In this way, the final obtained script list also includes the script text corresponding to the advertising insertion segment. Moreover, with the help of the large language model's ability to sort out the contextual semantics, this script text and the script text of the original business segments of the live broadcast business process can transition more smoothly in terms of natural semantics, presenting the audience with the feeling of advertising information being presented in natural conversation, playing a role similar to soft advertising, and further optimizing the user experience. Step S5300: Through a video generation service, the video server is invoked to generate a human-image broadcast video corresponding to each script text based on a template video. The template video reaches a preset duration, and each image frame contains a facial image captured from the same person. To achieve efficient operation, the computer program product of this application runs a video generation service, which communicates with the video server to drive the video server to generate corresponding human-image broadcast videos based on texts directly or indirectly submitted by the broadcaster user, such as script texts and inserted texts. After obtaining each script text from the script list, the digital broadcast control system of the broadcaster user's terminal device submits each script text to the video server in the network architecture through the video generation service. The video server invokes the template video preset by the broadcaster user and generates a human-image broadcast video corresponding to the script text based on the template video. Specifically, a voice-action driven model can be deployed in the video server. For example, this model first converts the text received by the video server, such as scripted text or inserted text, into a speech sequence using an acoustic model to obtain the corresponding audio data and determine the audio duration. Then, based on the audio duration, it randomly selects a video image frame sequence corresponding to the duration from the template video preset by the broadcaster. Next, based on temporal alignment, it uses the speech sequence to correct the mouth shape of the face image in each image frame of the video image frame sequence, obtaining a corrected image frame sequence. Finally, based on temporal alignment, the audio data and the corrected image frame sequence are encoded into a human-image spoken video. After the video server generates the human-image spoken video corresponding to each text, it can be pushed to the broadcaster's terminal device.The terminal device can then download the corresponding voice-over video to its local machine and establish a mapping with the corresponding script or interstitial text for subsequent use. Template videos can be pre-collected by the broadcaster and uploaded to a video server for storage. In this application, a pre-set regulated duration sufficient to ensure the diversity of human actions in the template video is implemented. The broadcaster collects source videos based on a reference duration determined by this regulated duration. After the broadcaster completes the collection of source videos, if the total duration of the source videos still does not reach the regulated duration, the terminal device or video server can use data augmentation to extend the source videos to a length that meets the regulated duration. Once the duration reaches the regulated duration, it can be used as a template video. HK 20115172 B Specification 10 In one embodiment, the broadcaster can preset multiple template videos. The characters in different template videos can be the same staff member or different staff members. Different template videos can have different styles, such as different spatial environments, different clothing styles, different desktop arrangements, etc. When the video server needs to generate a human-image lip-sync video, it randomly selects one of the template videos to capture a sequence of video image frames to generate the lip-sync video. When the broadcaster captures source video, it does so by recording upper-body images of the same person. The person in front of the camera may or may not speak, but can have various expressions and actions to enrich the action features in the same source video for final transfer to the template video. Following this requirement, all image frames in the source video contain facial images of the person; therefore, the template video also contains facial images of the same person, which can serve as the basis for voice-driven lip-syncing. Step S5400: Through the virtual camera driving service, the lip-sync videos corresponding to each script text in the script list are pushed to the e-commerce live streaming room according to the live streaming business process to implement the virtual live streaming activity. To further reduce the false alarm rate of the risk control system, the computer program product corresponding to the digital broadcast control system of this application can automatically install a virtual camera driver service on the terminal device of the broadcaster. After acquiring the script list and the corresponding human-image broadcast video, it calls the virtual camera driver service to push the human-image broadcast video corresponding to each script in the script list to the e-commerce live broadcast room one by one, thereby implementing the virtual live broadcast activity. Since the scripts in the activity list are organized in an orderly manner according to the various business links of the same live broadcast business process, the human-image broadcast video corresponding to each script can be called and pushed to the live broadcast server of the e-commerce platform according to the natural order of the scripts in the activity list. After receiving the video stream of the human-image broadcast video, the live broadcast server pushes it to the terminal devices of each viewer in the e-commerce live broadcast room.This allows each viewer to watch the live-streaming video and hear the corresponding scripted audio through the playback window on the corresponding page of the e-commerce live stream. Since the live-streaming video is driven by the scripted audio to correct lip movements, the voice and lip movements appear coordinated and natural to the viewer. It's clear that when a user starts a live stream to execute a virtual live stream activity, this application first retrieves the script list from the script server via a script generation service to obtain the scripted texts corresponding to each business step of the same live stream process. Then, a video generation service drives the video server to instantly generate live-streaming videos corresponding to each scripted text based on template videos that have reached a preset duration. Finally, a virtual camera driving service pushes these live-streaming videos into the e-commerce live stream to implement the virtual live stream activity. This demonstrates the advantages of this application in several aspects, including: In terms of dynamic recognition, template videos conforming to regulated durations can enrich and generalize the action features of facial images in the corresponding spoken video scripts, reducing the frequency of repeated action features detected by the risk control system; in terms of static recognition, the virtual camera-driven service avoids the inherent flaw of the risk control system mechanically judging robot broadcasting behavior based on data sources. As can be seen from HK 20115172 B Specification 11, through the combination of these two aspects, this application effectively prevents virtual live streaming activities from being mistakenly identified as robot broadcasting behavior by the e-commerce platform's risk control system, both in terms of dynamic and static recognition mechanisms. This reduces the frequency of the risk control system mistakenly triggering interference with the virtual live streaming activities in this e-commerce live streaming room, improves the stability and security of virtual live streaming activities, avoids unnecessary economic losses for the e-commerce store to which the live streaming room belongs, and also safeguards the application of virtual human live streaming in the e-commerce field, removing obstacles to its application. Based on any embodiment of the method in this application, referring to Figure 4, the virtual live streaming activity is implemented by pushing the corresponding human-image voiceover video of each script text in the script list to the e-commerce live streaming room according to the live streaming business process, including: Step S5410, responding to the dynamic insertion command triggered in the e-commerce live streaming room and determining the corresponding insertion text; At any time during the virtual live streaming activity, a dynamic insertion command can be triggered in the e-commerce live streaming room. The digital broadcast control system responds to the dynamic insertion command and determines the insertion text corresponding to the command. Of course, the insertion text is also a script text, which is used to generate the human-image voiceover video corresponding to the insertion text for insertion. For example: In one embodiment, the anchor user can trigger the dynamic insertion command through the digital broadcast control system in their terminal device. For example, the digital broadcast control system can provide a function button. After the anchor user touches the function button,Entering and submitting the insertion text triggers the dynamic insertion command; alternatively, the broadcaster can directly send a chat message to the e-commerce live stream, containing the input text to trigger the dynamic insertion command, thus determining the insertion text accordingly. These methods facilitate broadcasters inserting advertisements or other similar content into virtual live stream events. In another embodiment, viewers can also trigger the corresponding dynamic insertion command by sending chat messages in the e-commerce live stream. For example, a viewer might ask a question about a product and send a corresponding chat message containing the input text corresponding to the question. The digital broadcast control system can then match this input text to the appropriate reply text as the insertion text. This allows viewers to ask questions about products for sale in the e-commerce live stream, and the digital broadcast control system can then automatically respond, providing automated customer service. Step S5420: Through the video generation service, the video server is called to generate the human-image voiceover video corresponding to the inserted text based on the template video; Similar to step S5300, after the inserted text is determined, the digital broadcast control system submits the inserted text to the video server through its video generation service. After receiving the inserted text, the video server converts the inserted text into a speech sequence using an acoustic model, determines the corresponding video image frame sequence from the selected template video, and then calls the speech action driving model to correct the mouth shape of the face image in each image frame of the video image frame sequence according to the speech sequence, obtaining a corrected image frame sequence. Then, using the temporal alignment relationship, the speech sequence and the corrected image frame sequence are encoded into a human-image voiceover video and sent back to the video generation service. HK 20115172 B Specification 12 Step S5430: The human-image voiceover video corresponding to the inserted text is inserted after the human-image voiceover video being pushed to the e-commerce live broadcast room. After obtaining the corresponding voice-over video for the inserted text through its video generation service, the digital broadcast control system can insert it into the ongoing virtual live broadcast activity in the e-commerce live broadcast room. This allows the inserted voice-over video to be seamlessly integrated as the virtual live broadcast activity plays the voice-over videos corresponding to each script text in the script list according to the live broadcast business flow. Since the virtual live broadcast activity pushes the corresponding voice-over videos to the live broadcast server in real time according to a preset live broadcast business flow, i.e., the order of each script text in the script list, to maintain the smoothness of the virtual live broadcast activity, the system can first determine the voice-over video currently being pushed to the e-commerce live broadcast room (i.e., the live broadcast server), and then insert the voice-over video corresponding to the inserted text after the currently pushed voice-over video. Specifically, it can be inserted either immediately after the currently pushed voice-over video, or one or two voice-over videos after it.Specifically, the timing of the dynamic insertion command can be determined based on the response time. For example, in one embodiment, when the dynamic insertion command is used to answer questions asked by viewers via chat messages, it can be inserted immediately after the currently pushed voice-over video. In another embodiment, when the dynamic insertion command is used to insert advertisements, it can be inserted at a position several times later than the currently pushed voice-over video. Based on the above embodiments, it can be seen that during a virtual live streaming event, dynamic insertion commands can be triggered in various ways to determine the corresponding insertion text. Based on this, a corresponding voice-over video can be generated and inserted into a pre-determined script list for playback. This can be used for advertising integration and also for automatically providing automatic Q&A services to viewers. The entire process is relatively smooth, making the human-to-human interaction presented in the virtual live streaming event in the e-commerce live streaming room more realistic and natural, and significantly improving the user experience for viewers participating in the e-commerce live streaming room. Furthermore, since dynamic insertion commands can enrich the diversity of human-image video in the script list, it can further reduce the risk of virtual live streaming activities being identified as mechanically repetitive robot broadcasting behavior, that is, reduce the false trigger rate of the e-commerce platform's risk control system and ensure the stable operation of virtual live streaming activities. Based on any embodiment of the method of this application, please refer to Figure 5. Responding to the dynamic insertion command triggered in the e-commerce live streaming room, determining the corresponding insertion text includes: Step S5411, detecting user input information in the e-commerce live streaming room, performing intent recognition on the user input information to determine whether it carries an insertion intent; When it is necessary to identify dynamic insertion commands through chat messages in the e-commerce live streaming room, the digital broadcast control system of this application can be responsible for detecting each chat message in the public screen message flow in the e-commerce live streaming room, obtaining the chat messages of each user, including the anchor user and the audience user, that is, the user input information submitted by the corresponding user. Then, a pre-trained intent recognition model, as described in HK 20115172 B specification 13, is used to recognize the intent of the user input information to determine whether the user input message carries an insertion intent. When an insertion intent is detected, the corresponding dynamic insertion instruction can be triggered. Step S5412: When an insertion intent is detected, the dynamic insertion instruction corresponding to the insertion intent is triggered. The insertion type and input text corresponding to the insertion intent are determined based on the user input information. After recognizing that the user input information carries an insertion intent and triggering the corresponding dynamic insertion instruction, the digital broadcast control system can further determine the corresponding insertion type and input text based on the user input information and its implied insertion intent. For example, if the user input information submitted by the anchor is "Please help me generate an advertisement for XX model sports shoes", the intent recognition model can determine that there is an insertion intent corresponding to inserting an advertisement based on the user input information.After removing invalid characters and emoticons from the user input, the user's valid input text can be obtained. For example, if a viewer submits the input "Host, what material are your sneakers made of?", the intent recognition model can determine that there is an intention to insert a segment corresponding to answering a question. Similarly, after removing invalid characters and emoticons from the user input, the user's valid input text can be obtained. Step S5413: Determine the inferred text corresponding to the input text as the inserted text from the database corresponding to the inserted type. After determining the inserted type, the digital broadcast control system knows which interface to call. For example, for inserted advertisements, the interface provided by the advertising system can be called; for answering questions, the interface provided by the customer service system can be called. By calling the corresponding interface, the inserted text corresponding to the input text can be further obtained. For example, in an embodiment corresponding to interstitial advertising, the digital broadcast control system extracts product keywords from the user's input text, such as "XX model of sports shoes." Then, it submits this to an interface provided by the advertising system. The advertising system's interface, based on these product keywords, matches the target product corresponding to the keyword from the database of online stores belonging to the broadcaster on the e-commerce platform. It then obtains the product information of the target product, including but not limited to any one or more of the following: product images, product titles, product detail text, and product attribute data, depending on the preset business logic. This product information is then input into a preset advertising copy generation model to expand and generate plain text-based reasoning text, essentially generating the corresponding advertising copy, which is then used as the interstitial advertising text. As another example, in an embodiment corresponding to answering questions, the digital broadcast control system can directly submit user input text to the customer service system via a live call to its interface. This interface utilizes the semantic vector of the user input text to match semantically similar basic questions from the customer service system's database (question-answer database). It then retrieves a corresponding answer text from the question-answer database as the inferred text for the input text. The digital broadcast control system can then use this inferred text as the inserted text. The above embodiments demonstrate that the digital broadcast control system of this application possesses highly intelligent features. It can utilize user input information from e-commerce live streaming (HK 20115172 B specification 14) to perform intent recognition, determine the corresponding insertion intent and its input text, generate the inferred text corresponding to the input text by calling the interface corresponding to the insertion intent, and use the inferred text as the inserted text. The entire process is automated, eliminating the need for complex user operations. It automatically generates corresponding human-image voice-over videos based on user input information, and through the playback of these videos in virtual live streaming activities, it completes the response to user input.Creating the effect of a person in the live stream giving a speech while receiving a task assigned by the host user, or creating the effect of a person in the live stream responding to the needs of the audience user, can significantly improve the user experience. Furthermore, since the inserted text is dynamically generated based on user input, the content of each generated inserted text is generally different, resulting in different final human-image-based spoken video, further reducing the probability of accidentally triggering the e-commerce platform's risk control system. Based on any embodiment of the method in this application, inserting the human-image-based spoken video corresponding to the inserted text after the human-image-based spoken video being pushed to the e-commerce live stream includes: Step S5431, according to the order of business links in the live stream business process, sequentially loading the human-image-based spoken videos corresponding to each script text in the script list into the cache queue, and sequentially pushing the human-image-based spoken videos dequeued from the cache queue to the live stream server; In order to improve the memory efficiency of terminal devices and ensure the smoothness of virtual live stream activities at the data level, in this embodiment, the digital broadcast control system of this application can be further optimized by combining caching technology. Accordingly, the digital broadcast control system can load the voice-over videos corresponding to the script texts of each business segment in the script list of the live broadcast business process into a cache queue according to the order of each business segment in the preset live broadcast business process. Then, combined with the queue scheduling principle, the system controls each voice-over video to be dequeued from the cache queue in the order of its corresponding business segment. When a voice-over video is dequeued from the cache queue for consumption, the consumption thread is responsible for encoding and pushing the voice-over video to the live broadcast server. After receiving the corresponding video stream of the voice-over video, the live broadcast server decodes and re-encodes it and pushes it to the terminal devices of each viewer. Each viewer's terminal device decodes and plays the corresponding video stream, and can then watch the corresponding voice-over video. Step S5432: Detect and determine the head position of the voice-over video being dequeued in the cache queue, and insert the voice-over video corresponding to the inserted text into the next position after the head position. To insert the accompanying video message corresponding to the interstitial text into the existing live streaming workflow, caching technology can be used. First, identify the video messages currently being consumed and pushed out of the cache queue. Determine the position of this video message as the head of the cache queue. Then, insert the video message corresponding to the interstitial text into the next position after the head of the queue. This ensures that the video message immediately follows the video message being dequeued. Once the dequeued video message is consumed and the push is complete, the video message corresponding to the interstitial text can then be dequeued and consumed immediately.This ensures the immediacy of the playback of the video with the inserted text. HK 20115172 B Specification 15 As can be understood from the above embodiments, by combining caching technology to process the playback order between the video with the inserted text and the existing video with the inserted text in the live broadcast workflow, the playback of the video with the inserted text becomes a business step in the live broadcast workflow, playing naturally within it. Due to the use of caching technology, this process is smooth, the transmission is stable and fluid, and the digital broadcast control system exhibits good robustness. Furthermore, placing the video with the inserted text after the first position in the queue ensures its immediacy, and its efficiency advantage is significant when responding to user input. Based on any embodiment of the method in this application, referring to Figure 6, before obtaining the script list from the script server through the script generation service, the method includes: Step S4100, obtaining a material video with a duration of at least half of the regulated duration, wherein the material video contains facial images captured based on the same person; As disclosed above, the regulated duration in this application is preset to ensure sufficient diversity of human facial movements in the template video to effectively reduce the false trigger rate of the risk control system of the e-commerce platform. In this case, theoretically, when a template video needs to be generated, it is only necessary to record a template video of a specific person according to this regulated duration, but the time cost caused by recording the video in this way is high. In this case, when making a template video, a recording duration can be set, which can be any value greater than or equal to half of the regulated duration, in order to save recording time costs. Under this condition, the digital broadcast control system can open a recording program to record material videos of people. During recording, the camera can be aimed at the upper body of the person to ensure that the facial image of the person is present in each image frame of the material video. It should be noted that the source video does not require audio recording. Step S4200: Reverse the image frames in the source video to form its corresponding reverse video; After obtaining the source video, the digital broadcast control system reverses the image frames of the source video to transform it into a reverse video of the source video. The reverse video processed in this way, when played, visually enriches the diversity of the characters' movements, especially facial movements. Therefore, the reverse video is essentially a product of data enhancement of the source video in a superior way, thus eliminating the trouble of users recording long source videos. Step S4300: Splice the source video and its corresponding reverse video to form the template video used to create the human voiceover video. After determining the source video and the reverse video, the reverse video can be spliced ​​after the source video to achieve the splicing of the source video and its reverse video. The spliced ​​video...This allows it to be used directly as a template video. For easier future use, the template video can be associated with the corresponding identifier of its live streaming workflow and stored. Later, based on the identifier corresponding to the live streaming workflow, the appropriate template video can be used to create a voiceover video. It's easy to understand that the total duration of the resulting template video (HK 20115172 B, Manual 16) will inevitably reach the preset duration. Since the first frame of the reversed video is also the last frame of the source video, and the reversed video is spliced ​​after the source video, the movements of the characters in the entire template video are relatively smooth. Because the movements of the characters in the video frame sequence extracted from the middle of the template video will also be smooth. Similarly, since the last frame of the reversed video is also the first frame of the source video, when it's necessary to extract a video frame sequence at the end of the template video that exceeds the total duration of the template video, it can loop back to the first frame of the template video to continue extracting. The resulting video frame sequence will also inevitably have smooth movements of the characters. In some embodiments, if the total duration of the template video is much longer than the preset regulation duration, the template video can be cropped to control storage space. In this case, the duration to be cropped can be divided by two to obtain the duration to be deleted. Image frames corresponding to the duration to be deleted are then deleted from both the beginning and end of the template video. This ensures that the beginning and end image frames of the template video are responsive, and that the actions of the characters remain smooth when the video image frame sequence is repeatedly extracted based on the template video. As can be seen from the above embodiments, although the digital broadcast control system sets a regulation duration, the data augmentation method adopted in this embodiment can effectively reduce the duration of the recorded source video. By using the method of generating a reverse video based on the source video to expand the source video to obtain the template video, the time cost of recording the source video is reduced. At the same time, it can also ensure that when the video image frame sequence corresponding to the script text / inserted text is subsequently obtained based on the template video, the actions of the characters in the sequence are still relatively smooth and natural. This will not make the audience feel abrupt, nor will it easily trigger the alarm of the e-commerce platform's risk control system, thus improving the robustness of the virtual live broadcast event. Based on any embodiment of the method in this application, referring to Figure 7, before acquiring video material with a duration at least half of the regulated duration, the method includes: Step S3100: Calculating the interference rate corresponding to the interference by the e-commerce platform's risk control system when a virtual live streaming activity is executed in response to the virtual live streaming start command in the e-commerce live streaming room; During the process of a streamer user implementing virtual live streaming activities through the digital broadcast control system of this application for a long period of time, the digital broadcast control system can be responsible for recording the situation where each virtual live streaming activity triggers interference from the e-commerce platform's risk control system, and accordingly calculating the total number of virtual live streaming activities implemented, as well as the number of interferences corresponding to the interference events that trigger the risk control system interference, and dividing the number of interferences by the total number of implementations.This allows us to obtain the intervention rate corresponding to the intervention behavior of the risk control system. Intervention behaviors implemented by the risk control system include, but are not limited to, any punitive or warning-like actions such as time-limited suspension of broadcasting, immediate banning of broadcasting, deduction of points from the broadcaster, and sending alarm notifications to the broadcaster. Each such action is considered an intervention event, and the digital broadcast control system records it accordingly for subsequent statistical analysis. The digital broadcast control system determines the intervention rate when the broadcaster initiates a virtual live broadcast activity, triggering the corresponding virtual live broadcast start command (HK 20115172 B Manual 17), to ensure the smooth operation of the current virtual live broadcast activity. Step S3200: Determine whether the interference rate reaches a preset threshold. If it does, trigger a material video expansion command to update the regulated duration by adding a fixed duration. The updated regulated duration is then used as the recording duration. The digital broadcast control system also presets a threshold, which can be an empirical or experimental threshold. This threshold is used to judge the interference rate. When the interference rate reaches the threshold, a material video expansion command is triggered to guide the broadcaster to re-record a longer material video to generate a new template video. If the threshold is not reached, no further processing is required. The recommended threshold can be any value between 20% and 30%. That is, if relying on historical template videos for virtual live streaming results in 20% to 30% of the activities being interfered with by the risk control system, the user can be guided to recreate the template video. The increased interference rate of the risk control system may be due to two reasons. Firstly, the diversity of character movement features in the template video may not be sufficient to circumvent the dynamic detection mechanism's condition parameters. Secondly, the risk control system may have increased the requirements for these condition parameters. Regardless, the diversity of character movement features in the video can be ensured by extending the total duration of the template video. Accordingly, upon triggering the material video expansion command, this embodiment also updates the previously used regulatory duration. Specifically, a preset fixed duration can be used, and the original regulatory duration is added to this fixed duration to form the new regulatory duration. The material video is recorded using the new regulatory duration. This fixed duration can be a preset value, such as 5 minutes, 10 minutes, etc. Step S3300: Responding to the material video expansion command, the video recording program is started to record the material video using the updated regulatory duration for use in creating the template video. Upon triggering the video expansion command, the digital broadcast control system initiates a video recording program in response to the command. Using the updated specified duration, a new template video is created. Specifically, the new template video can be created following steps S3100 to S3300 in the previous embodiment. In this embodiment,The digital broadcast control system uses statistically obtained data on the interference rate of the risk control system in relation to virtual live streaming activities to intelligently decide whether to recreate the template video. When a high interference rate is detected, the system promptly extends the regulated duration by adding a fixed time limit and guides users to recreate template videos of the specified duration throughout the process. This ensures that the diversity of character action features in the template videos used reduces the interference rate of the risk control system in virtual live streaming activities, effectively reduces the false trigger rate of the risk control system, and improves the robustness of virtual live streaming activities. Based on any embodiment of the method in this application, referring to Figure 8, calling the video server to generate the corresponding human-image spoken video based on the template video / calling the video server to generate the corresponding human-image spoken video based on the template video includes: Step S6100: The video server obtains the spoken text / inserted text to be generated for the human-image spoken video, calls the acoustic model to generate the audio data of the spoken text / inserted text, and determines the corresponding spoken duration of the audio data; Since the inserted text is also a type of spoken text, the same principle applies to both. Therefore, in describing the process of the video server generating the human-image spoken video, this embodiment mainly focuses on the spoken text. Those skilled in the art should know that it is also applicable to the inserted text. After the video server obtains the spoken text or inserted text submitted by the digital broadcast control system for generating the human-image spoken video, it can call the pre-set acoustic model to convert the spoken text or inserted text into audio data. The audio data can be represented as a speech sequence for easy intermediate calling. The acoustic model can be any mature, known model, which can be directly implemented by those skilled in the art from existing technologies. Once the acoustic model generates corresponding audio data through its text-to-speech reasoning ability, the duration of the audio data is determined and can be used as the speech duration. Step S6200: The video server extracts a sequence of video image frames corresponding to the speech duration from the template video. The video server then, based on the speech duration, extracts a segment of image frames with a duration corresponding to the speech duration from the template video whose duration has reached the regulated duration, as a video image frame sequence corresponding to the speech duration, ensuring that the video image frame sequence maintains temporal alignment with the speech sequence of the audio data. When the video server extracts the video image frame sequence from the template video, it can do so by sequentially extracting image frames from different positions in the same template video to form corresponding video image frame sequences when processing different speech texts multiple times, or it can randomly locate and obtain the corresponding video image frame sequence from the template video for each processed speech text. Step S6300: The video server calls the voice action-driven model and generates audio data based on the corresponding script text / inserted text.The video server corrects the lip movements of facial images in the video image frame sequence to obtain a corrected image frame sequence. After obtaining the audio data and video image frame sequence corresponding to the speech duration, the video server can call its preset voice action driving model. This model can be any mature model, which can be flexibly selected by those skilled in the art. Using this model, the lip movements of facial images in each image frame of the video image frame sequence are corrected according to the voice data in the audio sequence, thereby achieving the correction of lip movements of facial images in the entire video image frame sequence, thus obtaining the corrected image frame sequence. In step S6400, the video server performs time alignment on the audio data and the corrected image frame sequence to generate a human-image voice-over video corresponding to the speech text / inserted text. After obtaining the corrected image frame sequence, the video server encodes the corrected image frame sequence and the audio sequence representing the audio data according to the time alignment relationship to generate a human-image voice-over video corresponding to the speech text or inserted text. The video server pushes the human-image voice-over video to the digital broadcast control system in the terminal device of the broadcaster user, which can then be used to implement virtual live broadcast activities. HK 20115172 B Specification 19 In this embodiment, the video server is responsible for centrally processing the generation of human-image broadcast videos corresponding to various scripted texts. Its business logic is centralized and reusable, and its computational efficiency is high. Since the video server can use the acoustic model corresponding to text-to-speech and the speech-action correction model to extract material from template videos to generate human-image broadcast videos corresponding to scripted texts, the quality of the generated human-image broadcast videos is high. When its human-image broadcast videos are used in virtual live streaming activities, it helps to reduce the false trigger rate of the risk control system of e-commerce platforms by leveraging the diverse characteristics of human actions in the human-image broadcast videos. Referring to Figure 9, another embodiment of this application also provides a live streaming risk control and accidental touch prevention device, which includes a live streaming response module 5100, a script acquisition module 5200, a video acquisition module 5300, and a live streaming module 5400. The live streaming response module 5100 is configured to respond to a virtual live streaming start command and start an e-commerce live streaming room on an e-commerce platform for executing virtual live streaming activities. The script acquisition module 5200 is configured to obtain a script list from a script server through a script generation service. The script list contains script texts corresponding to different business stages in the same live streaming business process. The video acquisition module 5300 is configured to call a video server through a video generation service to generate a human-image voiceover video corresponding to each script text based on a template video. The template video reaches a preset regulated duration, and each image frame contains a facial image captured based on the same person. The live streaming module 5400 is configured to use a virtual camera driver service.According to the live streaming business process, the corresponding human-image spoken video for each text in the script list is pushed to the e-commerce live streaming room to implement the virtual live streaming activity. Based on any embodiment of the device in this application, the live streaming module 5400 includes: an insertion response module, configured to respond to a dynamic insertion command triggered in the e-commerce live streaming room and determine the corresponding insertion text; an insertion generation module, configured to call a video server based on the template video to generate the human-image spoken video corresponding to the insertion text through a video generation service; and a video insertion module, configured to insert the human-image spoken video corresponding to the insertion text after the human-image spoken video being pushed to the e-commerce live streaming room. Based on any embodiment of the device in this application, the insertion response module includes: an input detection module, configured to detect user input information in the e-commerce live streaming room, perform intent recognition on the user input information to determine whether it carries an insertion intent; an information extraction module, configured to trigger a dynamic insertion instruction corresponding to the insertion intent when it carries an insertion intent, and determine the insertion type and input text corresponding to the insertion intent based on the user input information; and a text determination module, configured to determine the inferred text corresponding to the input text from the database corresponding to the insertion type as the insertion text. Based on any embodiment of the device in this application, the video insertion module includes: a cache scheduling module, configured to sequentially load the human-image voice-over videos corresponding to each script text in the script list into a cache queue according to the order of business links in the live broadcast business process, and push the human-image voice-over videos sequentially dequeued from the cache queue to the live broadcast server; and a positioning insertion module, configured to detect and determine the head position of the human-image voice-over video being dequeued and pushed in the cache queue, and insert the human-image voice-over video corresponding to the insertion text into the position after the head position. Based on any embodiment of the device in this application, prior to the operation of the script acquisition module 5200, the live streaming risk control and accidental touch prevention device of this application includes: a material acquisition module, configured to acquire material videos with a duration of at least half of the regulated duration, the material videos including facial images captured based on the same person; a reverse playback expansion module, configured to arrange the image frames in the material video in reverse order to form its corresponding reverse playback video; a template creation module, configured to splice the material video and its corresponding reverse playback video to form the template video used to create the portrait voiceover video. Based on any embodiment of the device in this application, prior to the operation of the material acquisition module, the live streaming risk control and accidental touch prevention device of this application includes: an interference statistics module, configured to calculate the interference rate corresponding to the interference of the risk control system of the e-commerce platform in the e-commerce live streaming room when the virtual live streaming activity is executed in response to the virtual live streaming start command; a duration update module,The system is configured to determine whether the interference rate reaches a preset threshold. When the preset threshold is reached, a material video expansion command is triggered to update the regulation duration by adding a fixed duration. The updated regulation duration is then used as the recording duration. The video re-recording module is configured to respond to the material video expansion command, start the video recording program, and record the material video using the updated regulation duration for use in creating the template video. Based on any embodiment of the apparatus in this application, the video acquisition module 5300 / the insertion generation module includes: a generation preparation module, configured to have a video server acquire the script text / insertion text of the human image speech video to be generated, call an acoustic model to generate audio data of the script text / insertion text, and determine the corresponding speech duration of the audio data; a template extraction module, configured to have a video server extract a video image frame sequence corresponding to the speech duration from the template video; a lip-shape correction module, configured to have a video server call a voice action driving model, and correct the lip-shape action of the face image in the video image frame sequence according to the audio data generated corresponding to the script text / insertion text, to obtain a corrected image frame sequence; and a video generation module, configured to have the video server perform time-series alignment of the audio data and the corrected image frame sequence, and generate the human image speech video corresponding to the script text / insertion text. Based on any embodiment of this application, please refer to Figure 10. Another embodiment of this application also provides a computer device, as shown in Figure 10, which is a schematic diagram of the internal structure of the computer device. The computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected via a system bus. The computer device includes a computer-readable storage medium storing an operating system, a database, and a computer program encapsulated with computer-readable instructions. The database may store a sequence of control information. When the computer-readable instructions are executed by the processor, the processor can implement a live-streaming risk control method for preventing accidental touches. The processor of the computer device provides computing and control capabilities to support the operation of the entire computer device. The memory of the computer device may store computer-readable instructions. When these computer-readable instructions are executed by the processor, the processor can execute the live-streaming risk control method for preventing accidental touches as described in this application. The network interface of the computer device is used for communication with a terminal. Those skilled in the art will understand that the structure shown in Figure 10 is merely a block diagram of a portion of the structure related to the solution in this application (HK 20115172 B Specification 21) and does not constitute a limitation on the computer device to which the solution in this application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In this embodiment, the processor is used to execute the specific functions of each module and its sub-modules in Figure 9.The memory stores the program code and various data required to execute the above modules or sub-modules. The network interface is used for data transmission between the user terminal and the server. In this embodiment, the memory stores the program code and data required to execute all modules / sub-modules in the live streaming risk control and accidental touch prevention device of this application. The server can call the server's program code and data to execute the functions of all sub-modules. This application also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, they cause one or more processors to perform the steps of the live streaming risk control and accidental touch prevention method described in any embodiment of this application. This application also provides a computer program product, including a computer program / instructions. When the computer program / instructions are executed by one or more processors, they implement the steps of the live streaming risk control and accidental touch prevention method described in any embodiment of this application. Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a computer-readable storage medium such as a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM). The above description is only a partial embodiment of this application. It should be noted that those skilled in the art can make several improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application. In summary, this application effectively prevents virtual live streaming activities from being mistakenly identified as robot broadcasting behavior by the risk control system of e-commerce platforms, both from a dynamic and static identification mechanism perspective. This reduces the frequency of the risk control system's erroneous intervention in the virtual live streaming activities of this e-commerce live streaming room, improves the stability and security of virtual live streaming activities, avoids unnecessary economic losses to the e-commerce stores to which the live streaming room belongs, and also safeguards the application of virtual human live streaming in the e-commerce field, removing obstacles to its application. HK 20115172 B 1 Claim 1: A method for preventing accidental touches during live streaming risk control, characterized in that it includes: responding to a virtual live streaming start command and starting an e-commerce live streaming room on an e-commerce platform for executing virtual live streaming activities; obtaining a script list from a script server through a script generation service, wherein the script list contains script texts corresponding to different business stages in the same live streaming business process; obtaining a video material with a duration of at least half of a preset regulated duration, wherein the video material contains facial images collected based on the same person, and reversing the image frames in the video material to form a corresponding reverse video.The source video is stitched together with the corresponding reverse video to form a template video that meets the specified duration. A video generation service is used to call a video server to generate a human-image-based spoken video corresponding to each script text in the script list based on the template video. When generating each human-image-based spoken video, the video server randomly selects and extracts a sequence of video image frames corresponding to the script duration from the template video. A virtual camera driver service is installed on the terminal device that generates the human-image-based spoken video. Through the virtual camera driver service, the human-image-based spoken video is pushed to the e-commerce live streaming room according to the live streaming business process to implement the virtual live streaming activity. The interference rate of the virtual live streaming activity due to interference from the risk control system is calculated, and when the interference rate reaches a preset threshold, the duration of the specified limit is extended and the template video is remade. 2. The live streaming risk control method for preventing accidental touches according to claim 1, characterized in that, pushing the human-image voiceover video to the e-commerce live streaming room to implement the virtual live streaming activity according to the live streaming business process includes: responding to a dynamic insertion command triggered in the e-commerce live streaming room and determining the corresponding insertion text; calling the video server to generate a human-image voiceover video corresponding to the insertion text based on the template video through the video generation service; inserting the human-image voiceover video corresponding to the insertion text after the human-image voiceover video being pushed to the e-commerce live streaming room. 3. The live streaming risk control method for preventing accidental touches according to claim 2, characterized in that, responding to a dynamic insertion command triggered in the e-commerce live streaming room and determining the corresponding insertion text includes: detecting user input information in the e-commerce live streaming room, performing intent recognition on the user input information to determine whether it carries an insertion intent; when carrying an insertion intent, triggering a dynamic insertion command corresponding to the insertion intent, determining the insertion type and input text corresponding to the insertion intent according to the user input information; determining the inferred text corresponding to the input text from the database corresponding to the insertion type as the insertion text. 4. The live streaming risk control method for preventing accidental touches according to claim 2, characterized in that, inserting the HK 20115172 B 2 human voiceover video corresponding to the inserted text after the human voiceover video being pushed to the e-commerce live streaming room includes: according to the order of business links in the live streaming business process, orderly loading the human voiceover videos corresponding to each script text in the script list into the cache queue, and pushing the human voiceover videos orderly dequeued from the cache queue to the live streaming server; detecting and determining the head position of the human voiceover video being dequeued in the cache queue, and inserting the human voiceover video corresponding to the inserted text into the position after the head position. 5. The live streaming risk control method for preventing accidental touches according to claim 1, characterized in that,The step of calling the video server to generate a human-image spoken video corresponding to each script text in the script list based on the template video includes: the video server acquiring the script text, calling an acoustic model to generate audio data of the acquired script text, and determining the corresponding script duration of the audio data; the video server randomly locating and extracting a video image frame sequence corresponding to the script duration from the template video; the video server calling a voice action driving model to correct the mouth movements of the facial images in the video image frame sequence according to the audio data, obtaining a corrected image frame sequence; and the video server performing time-series alignment of the audio data and the corrected image frame sequence to generate a human-image spoken video corresponding to the script text. 6. The live streaming risk control and accidental touch prevention method according to claim 5, characterized in that, before performing time-series alignment of the audio data and the corrected image frame sequence, it further includes: if the duration of the template video is greater than the regulated duration, dividing the duration to be cut by two to obtain the duration to be deleted, and deleting the image frames corresponding to the duration to be deleted from both the beginning and end of the template video. 7. The method according to claim 1, wherein obtaining the script list from the script server comprises: submitting basic material text and prompt text to the script server; and having the script server invoke a neural network model to generate the script list matching the live streaming business process based on the basic material text and prompt text. 8. A live streaming risk control and accidental touch prevention device, characterized in that it comprises: a live streaming response module, configured to respond to a virtual live streaming start command and start an e-commerce live streaming room in an e-commerce platform for executing virtual live streaming activities; a script acquisition module, configured to obtain a script list from a script server through a script generation service, wherein the script list contains script texts corresponding to different business links in the same live streaming business process; a material acquisition module, configured to acquire a material video with a duration of at least half of a preset regulation duration, wherein the material video contains facial images collected based on the same person; a reverse playback expansion module, configured to arrange the image frames in the material video in reverse order to form a corresponding reverse playback video; and a template creation module, configured to splice the material video and the corresponding reverse playback video to form a template video that reaches the regulation duration; The video acquisition module is configured to use a video generation service to call a video server to generate corresponding human-image spoken videos for each speech text in the speech script list based on the template video. When generating each human-image spoken video, the video server randomly selects and extracts a sequence of video image frames corresponding to the speech script duration from the template video. The live streaming module is configured to use a virtual camera driver service installed on the terminal device that generates the human-image spoken videos.The live streaming business process is described above. The human-image voiceover video is pushed to the e-commerce live streaming room to implement the virtual live streaming activity. An interference statistics module is configured to count the interference rate of the virtual live streaming activity caused by the risk control system, such that when the interference rate reaches a preset threshold, the duration update module is triggered to extend the regulated duration, and the video re-recording module is triggered to recreate the template video. 9. A computer device, comprising a central processing unit and a memory, characterized in that the central processing unit is used to call and run a computer program stored in the memory to perform the steps of the method as described in any one of claims 1 to 7. 10. A computer program product, characterized in that it comprises a computer program / instructions, which, when run by a processor, perform the steps of the method as described in any one of claims 1 to 7. HK 20115172 B Instruction Manual Appendix 1 / 5 Figure 1 Live Streaming Equipment Script Server Live Streaming Server Video Server Figure 2 81 82 80 HK 20115172 B Instruction Manual Appendix 2 / 5 Responding to the virtual live streaming start command, the e-commerce live streaming room used to execute the virtual live streaming activity is started in the e-commerce platform. Through the script generation service, the script list is obtained from the script server. The script list contains script texts corresponding to different business links in the same live streaming business process. Through the video generation service, the video server is called to generate human image voice-over videos corresponding to each script text based on the template video. The template video reaches the preset regulation duration, and each of its image frames contains facial images captured based on the same person. S5300 S5200 S5100 Through the virtual camera driver service, the human image voice-over videos corresponding to each script text in the script list are pushed to the e-commerce live streaming room according to the live streaming business process to implement the virtual live streaming activity. S5400 Figure 3 In response to a dynamic insertion command triggered in the e-commerce live stream, the corresponding insertion text is determined. Through a video generation service, a video server is invoked to generate a human-image voiceover video corresponding to the insertion text based on the template video. This human-image voiceover video is then inserted after the human-image voiceover video being pushed to the e-commerce live stream. (S5430 S5420 S5410 Figure 4 HK 20115172 B Specification Figure 3 / 5) User input information in the e-commerce live stream is detected, and intent recognition is performed on the user input information to determine whether it carries an insertion intent. If it does, a dynamic insertion command corresponding to that intent is triggered.Based on the user input information, determine the insertion type and input text corresponding to the insertion intention. Determine the inference text corresponding to the input text from the database corresponding to the insertion type as the insertion text. S5413 S5412 S5411 Figure 5. Obtain a material video with a duration of at least half of the regulated duration. The material video contains facial images collected based on the same person. Reverse the image frames in the material video to form its corresponding reverse video. Splice the material video and its corresponding reverse video to form the template video used to create the human voiceover video. S4300 S4200 S4100 Figure 6 HK 20115172 B Specification Figure 4 / 5. Statistically analyze the interference rate corresponding to the interference of the e-commerce platform's risk control system in the e-commerce live broadcast room when the virtual live broadcast activity is executed in response to the virtual live broadcast start command. Determine whether the interference rate reaches a preset threshold. When the preset threshold is reached, trigger the material video expansion command to update the regulated duration by adding a fixed duration. The updated regulation duration is used as the recording duration in response to the material video expansion instruction. The video recording program is started to record the material video using the updated regulation duration for the production of the template video S3300 S3200 S3100 Figure 7 The video server obtains the script text / interlude text to be generated for the human voice-over video, calls the acoustic model to generate the audio data of the script text / interlude text, and determines the corresponding script duration of the audio data. The video server extracts the video image frame sequence corresponding to the script duration from the template video. The video server calls the voice action driving model, and corrects the mouth movements of the face images in the video image frame sequence according to the audio data generated corresponding to the script text / interlude text, to obtain the corrected image frame sequence S6300 S6200 S6100. After the video server aligns the audio data and the corrected image frame sequence in time, it generates the human voice-over video corresponding to the script text / interlude text S6400 Figure 8 HK 20115172 B Specification Appendix 5 / 5 Live Streaming Response Module, Script Acquisition Module, Video Acquisition Module 5100 5200 5300, Push Streaming Module 5400. Figure 9: Computer Equipment, Processor, System Bus, Network Interface, Storage Media, Computer Program, Memory, Operating System. Figure 10: HK 20115172 B.

Images