Live broadcast method, device and terminal of virtual digital object

Abstract

A method and device for live broadcasting of a virtual digital object, and a terminal, the method comprising: obtaining user behavior information in the process of driving the live broadcasting of the virtual digital object according to the explanation content, wherein the explanation content includes multiple content nodes; after the first content node is explained, determining the response content of the current response node according to the user behavior information in the current response interval, wherein the first content node is any one of the multiple content nodes; and driving the live broadcasting of the virtual digital object according to the response content. Through the scheme provided by the present application, dynamic interaction with the user can be achieved during the explanation process of the virtual digital object.

Description

Technical Field

[0001] The present application relates to the technical field of virtual digital objects, and in particular to a method, device and terminal for live broadcasting of virtual digital objects. Background Art

[0002] With the development of virtual reality, augmented reality and rendering technologies, virtual live broadcast technology has emerged. Virtual live broadcast technology refers to the technology that uses virtual digital objects (for example, virtual digital people) as virtual anchors to replace real anchors for live broadcasts. Compared with real anchors, virtual anchors can broadcast live for a long time, which has obvious advantages.

[0003] During the live broadcast, users can send bullet screens and comments to interact with the live broadcaster, and the live broadcaster will manually select a comment to reply. In the scenario where the virtual broadcaster broadcasts live, there is currently no relevant solution for how the virtual broadcaster interacts with users watching the live broadcast. Summary of the invention

[0004] One of the technical objectives of the present application is to provide a live broadcast method of a virtual digital object, which can realize dynamic interaction between a virtual anchor and a user during the virtual live broadcast process.

[0005] To solve the above technical problems, an embodiment of the present application provides a method for live broadcasting of a virtual digital object, the method comprising: obtaining user behavior information in the process of driving the live broadcasting of the virtual digital object according to explanation content, wherein the explanation content comprises multiple content nodes; after the first content node is explained, determining the response content of the current response node according to the user behavior information in the current response interval, the first content node being any one of the multiple content nodes; and driving the live broadcasting of the virtual digital object according to the response content.

[0006] Optionally, the end time of the current response interval is the end time of a previous node; wherein the previous node is the first content node, or the previous node is a previous response node.

[0007] Optionally, if there is a response node between the first content node and the second content node, the starting time of the current response interval is the starting time of the previous response node of the first content node, and the second content node is the previous content node of the first content node; if there is no response node between the first content node and the second content node, the starting time of the current response interval is the starting time of the first content node.

[0008] Optionally, if there is a response node of the same type between the first content node and the second content node, the starting time of the current response interval is the starting time of the previous response node of the same type of the first content node, and the second content node is the previous content node of the first content node; if there is no response node of the same type between the first content node and the second content node, the starting time of the current response interval is the starting time of the first content node; wherein, the response node of the same type is a response node of the same type as the current response node.

[0009] Optionally, the current response node is an interactive node, which is used to respond to the user's interactive behavior. Determining the response content of the current response node includes: determining the highest priority and / or latest interactive behavior in the current response interval, recording it as the behavior to be responded to; and determining the response content of the current response node based on the behavior to be responded to.

[0010] Optionally, the current response node is a question-and-answer node, which is used to answer questions raised by users. Determining the response content of the current response node includes: selecting a question to be answered from a set of candidate questions, wherein the set of candidate questions includes standard questions hit within the current response interval, and the question to be answered has an unanswered label; and determining the answer to the question to be answered.

[0011] Optionally, if the current response node is the first question-and-answer node after the first content node, each standard question in the candidate question set has the unanswered label.

[0012] Optionally, the explanation content includes multiple content groups, each content group includes at least one content node, and the method further includes: when driving the virtual digital object to explain to the third content node, judging whether to insert a response group after the third content node, the response group is one of the multiple content groups, or is another group other than the multiple content groups, and the third content node is the last content node in any content group; if the judgment result is yes, then after the third content node is explained, driving the virtual digital object to explain the response group.

[0013] Optionally, determining whether to insert a response group after the third content node includes: determining a standard sentence that is hit within a group response interval, wherein the starting time of the group response interval is the starting time of a fourth content node, and the ending time is the starting time of the third content node, and the fourth content node is the last content node in a content group before the current content group; determining whether there is a standard sentence among the hit standard sentences that is hit more than a preset threshold of the group, and if so, inserting the response group.

[0014] In order to solve the above-mentioned technical features, an embodiment of the present application also provides a live broadcast device for a virtual digital object, the device comprising: an acquisition module, used to acquire user behavior information in the process of driving the live broadcast of the virtual digital object according to the explanation content, wherein the explanation content includes multiple content nodes; a response determination module, used to determine the response content of the current response node according to the user behavior information in the current response interval after the first content node is explained, the first content node being any one of the multiple content nodes; a driving module, used to drive the live broadcast of the virtual digital object according to the response content.

[0015] The embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned live broadcast method of the virtual digital object are executed.

[0016] An embodiment of the present application also provides a terminal, including a memory and a processor, wherein the memory stores a computer program that can be executed on the processor, and the processor executes the steps of the above-mentioned live broadcast method of virtual digital objects when running the computer program.

[0017] Compared with the prior art, the technical solution of the embodiment of the present application has the following beneficial effects:

[0018] In the scheme of the embodiment of the present application, in the process of driving the live broadcast of the virtual digital object according to the explanation content, the user behavior information is obtained, and the explanation content includes multiple content nodes. After the first content node is explained, the response content of the current response node can be determined according to the user behavior information in the current response interval, and the virtual digital object can be driven according to the response content. In the scheme of the embodiment of the present application, the explanation content has a structured feature, and the response node between two adjacent content nodes responds to the user behavior information. Since the current response interval is dynamically adjusted for the response node, it is possible to dynamically respond to user behavior in different response time intervals. In this way, it is possible to dynamically interact with the audience during the process of the virtual digital object live broadcasting the pre-made content. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a schematic diagram of an application scenario of a live broadcast method of a virtual digital object in an embodiment of the present application;

[0020] Figure 2 It is a flowchart of a method for live broadcasting of a virtual digital object in an embodiment of the present application;

[0021] Figure 3 is a schematic diagram of a content node in an embodiment of the present application;

[0022] Figure 4 It is a flowchart of another method for live broadcasting of a virtual digital object in an embodiment of the present application;

[0023] Figure 5 is a schematic diagram of a content group in an embodiment of the present application;

[0024] Figure 6 It is a structural schematic diagram of a live broadcast device for a virtual digital object in an embodiment of the present application. DETAILED DESCRIPTION

[0025] As described in the background art, in a scenario where a virtual anchor is broadcasting live, there is currently no relevant solution for how the virtual anchor interacts with users who are watching the live broadcast.

[0026] In the scheme of the embodiment of the present application, in the process of driving the live broadcast of the virtual digital object according to the explanation content, user behavior information is obtained, and the explanation content includes multiple content nodes. After the first content node is explained, the response content of the current response node can be determined according to the user behavior information in the current response interval, and the live broadcast of the virtual digital object can be driven according to the response content. In the scheme of the embodiment of the present application, the explanation content has a structured feature, and the response node between two adjacent content nodes responds to the user behavior information. Since the current response interval is dynamically adjusted for the response node, it is possible to dynamically respond to user behavior in different response time intervals. In this way, it is possible to dynamically interact with the audience during the process of the virtual digital object live broadcast explaining the pre-made content.

[0027] In order to make the above-mentioned purposes, features and beneficial effects of the present application more obvious and easy to understand, the specific embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0028] Reference Figure 1 , Figure 1 This is a schematic diagram of an application scenario of a live broadcast method of a virtual digital object in an embodiment of the present application. Figure 1 The application scenarios of the embodiments of the present application are described in a non-limiting manner.

[0029] As described above, the application scenario of the embodiment of the present application can be a virtual live broadcast scenario. Specifically, the B-end user can pre-set the content to be explained, and then drive the virtual digital object to explain the pre-set content. Among them, the B-end user refers to the user who sets the explanation content of the virtual digital object. The virtual digital object can be a virtual digital person of various images, for example, it can be a virtual teacher, a virtual anchor, etc. The B-end user can pre-set the content to be explained according to actual needs. The content can be a product introduction, a teaching content explanation, etc., and this embodiment does not limit this. For the convenience of description, the pre-set content will be recorded as the explanation content below.

[0030] In a specific implementation, the explanation content set by the B-side user can be text content or voice content, and this embodiment does not limit this.

[0031] In the solution of the embodiment of the present application, the explanation content has a structured feature. Figure 1 As shown, the explanation content may include multiple content groups (e.g., Figure 1 In the process of virtual live broadcast, the virtual digital object can be driven to explain the multiple content groups according to the preset group order. The group order can be used to indicate the explanation order of the content groups.

[0032] Furthermore, each content group may include multiple content nodes ( Figure 1 Not shown), each content node has corresponding content, and the content corresponding to each content node is at least a part of the explanation content. For each content group, the content corresponding to each content node in the content group can be explained according to the pre-set node order.

[0033] In a non-limiting example, the explanation content is related to the product, that is, the B-side user sets the explanation content to drive the virtual digital object to explain the specific product. More specifically, the explanation content may include product content and general content. For example, any content group may be a product content group or a general content group. The content corresponding to each content node in the product content group may be a product introduction, and the content corresponding to each node in the general content group may be other content besides the product introduction, for example, it may be popular science content such as instructions for use.

[0034] As can be seen from the above, in the solution of this embodiment, when the virtual digital object is driven to perform live explanation of the explanation content, the explanation can be performed according to the pre-set explanation order. On the one hand, the content group currently explained is determined, and the content corresponding to each content node in the content group is explained according to the node order. When the last content node in the content group currently explained is explained, the next content group to be explained is determined according to the group order. As mentioned above, the explanation order connects the content corresponding to each content node of each content group in series, which can form a complete and clearly structured explanation content.

[0035] It should be noted that, in this embodiment, "the explanation of the content node is completed" means that the content corresponding to the content node is completed. The process of explaining according to the preset explanation order described above can be called the "main line process".

[0036] Furthermore, during the live broadcast, multiple content groups will be repeatedly played in rotation. Specifically, after the contents corresponding to the multiple content groups are explained in the explanation order, you can return to the first content group and continue to explain in the explanation order, and so on.

[0037] In a specific implementation, the B-side user may also pre-set a branch process (for example, Figure 1 In the branch flow 1 and branch flow 2 shown, the branch flow may have a trigger condition, and when the trigger condition is met, the branch flow may be triggered to enter. The trigger condition may be any one or a combination of the following: a timing trigger condition and a number of trigger conditions.

[0038] Specifically, the timing trigger condition may mean that when a predetermined time is reached, a branch process may be entered. If the virtual digital object is being driven to explain a certain content group when the predetermined time is reached, the branch process may be entered after the current content group is explained.

[0039] Furthermore, the number of trigger conditions may mean that when the number of C-end users in the live broadcast room reaches a preset number threshold, the branch process can be entered. Among them, the C-end user in the embodiment of the present application refers to the user who watches the live broadcast of the virtual digital object. In other words, the C-end user may refer to the audience watching the live broadcast.

[0040] In the main process, the virtual digital object can also be driven to interact with the C-end user. Interacting with the C-end user may include: replying to questions raised by the C-end user, and / or responding to interactive operations issued by the C-end user, etc., but is not limited to these. Figure 1 As shown, the virtual digital object can be driven to answer questions after content group 4 is explained, answer questions after content group 5 is explained, and drive the virtual digital object to respond interactively after content group 6 is explained.

[0041] For details about driving the virtual digital object to interact with the C-end user in the main process, please refer to the specific description below.

[0042] It should be noted that, in this embodiment, driving the virtual digital object may be driving the virtual digital object to generate multimodal information. Multimodal information may include information of multiple modes, including but not limited to: expressions, voices, and actions of the virtual digital object, as well as displayed text, displayed images, displayed videos, other audio, etc. The method for generating multimodal information may be various existing appropriate methods, and this embodiment does not limit this.

[0043] Reference Figure 2 , Figure 21 is a flowchart of a method for live broadcasting a virtual digital object in an embodiment of the present application. The method can be executed by a terminal, and the terminal can be any existing terminal device with data receiving and data processing capabilities, such as a mobile phone, a computer, an Internet of Things device, and a server, but is not limited thereto. In a specific example, the terminal can be a device of a B-end user, but is not limited thereto. In other examples, Figure 2 The method shown can be collaboratively executed by a B-end user's device and a natural language processing (NLP) platform. Figure 2 The illustrated method for live broadcasting of a virtual digital object may include the following steps S21 to S23.

[0044] Step S21: acquiring user behavior information in the process of driving the live broadcast of the virtual digital object according to the explanation content, wherein the explanation content includes a plurality of content nodes;

[0045] Step S22: after the first content node is explained, the response content of the current response node is determined according to the user behavior information in the current response interval;

[0046] Step S23: driving the live broadcast of the virtual digital object according to the response content.

[0047] It can be understood that, in a specific implementation, the method can be implemented in the form of a software program, which runs in a processor integrated inside a chip or a chip module; alternatively, the method can be implemented in the form of hardware or a combination of hardware and software, such as using a dedicated chip or chip module, or using a dedicated chip or chip module in combination with a software program.

[0048] In the specific implementation of step S21, the user behavior information is obtained in the process of driving the virtual digital object according to the explanation content. In other words, the user behavior information can be obtained in the process of the virtual digital object being broadcast live.

[0049] Specifically, user behavior information may record data on interactive behaviors initiated by C-end users when watching live broadcasts. More specifically, user behavior information may include time information and content information, wherein time information may be used to indicate the moment when the user initiated the interactive behavior, and content information may be used to indicate the type or content of the interactive behavior. The types of interactive behaviors may include problematic behaviors and interactive behaviors, wherein problematic behaviors may refer to asking questions, wherein interactive behaviors may include: sharing a live broadcast room, following a live broadcast room, entering a live broadcast room, giving likes, etc. Further, content information may be any of the following: questions raised by C-end users when watching live broadcasts, comments issued by C-end users when watching live broadcasts, interactive behaviors initiated by C-end users, etc., but is not limited thereto.

[0050] It should be noted that in the solution of this embodiment, user behavior information can be obtained during the entire virtual live broadcast process. Specifically, user behavior information can be obtained during the process of driving the virtual digital object to explain the content corresponding to the content node, and user behavior information can also be obtained during the process of driving the virtual digital object to interact with the user.

[0051] In a specific example, the B side can obtain user behavior information and send it to the NLP platform for analysis, obtain the analysis results, and send the analysis results to the device of the B side user. The NLP platform's analysis process of user behavior information can include: determining whether the question raised by the user hits the standard question and which standard question it hits, and / or determining whether the comment issued by the user hits the standard sentence and which standard sentence it hits, etc. For more information about hitting the standard question and hitting the standard sentence, please refer to the relevant description below.

[0052] In the specific implementation of step S22, after the first content node is explained, and before the next content node of the first content node is explained, it can be determined whether it is necessary to respond to the user's interactive behavior, and when it is determined to respond to the interactive behavior, the virtual digital object is driven to respond. Wherein, the first content node is any one of the multiple content nodes. In other words, between two adjacent content nodes, it can be determined whether it is necessary to respond to the user's interactive behavior, and when it is determined to respond to the interactive behavior, the virtual digital object is driven to respond. More specifically, the first content node can be any one of the content nodes in the main line process. In other embodiments, the first content node can also be a content node in the response group described below.

[0053] In a specific implementation, the B-side user can pre-set a response threshold M, which is used to indicate the maximum number of responses to user interactions between two adjacent content nodes. In other words, after the first content node is explained, there are M opportunities to respond to the user's interactive behavior, and these M opportunities can be called M response opportunities. For the i-th response opportunity, the current response interval can be determined to determine whether to use the i-th response opportunity to respond based on the user behavior information in the current response interval. If the judgment result is yes, a response node can be inserted, and the virtual digital object can be driven to respond at the inserted response node.

[0054] Specifically, it can be determined whether to insert a response node based on the user behavior information in the current response interval. The response node to be inserted can be understood as the current response node, and the current response interval can be the response interval corresponding to the current response node. Figure 3 The start time and end time of the current response interval are described in detail.

[0055] The first aspect is about the starting time of the current response interval.

[0056] In the first example, the start time of the current response interval is the start time of the first content node. Figure 3 As shown, the starting time of the current response interval is the starting time t1 of the first content node. By adopting such a scheme, the starting time of the most recent content node is used as the starting time of the current response interval, which is conducive to ensuring that the acquired user behavior information has good real-time performance, and can make the interaction between the virtual digital object and the user revolve around the content corresponding to the first content node, so that the response of the virtual digital object has good real-time performance and relevance.

[0057] In the second example, if there is a response node between the second content node and the first content node, the start time of the current response interval is the start time of the previous response node of the first content node. If there is no response node between the second content node and the first content node, the start time of the first content node is the start time of the current response interval. The second content node and the first content node are two adjacent content nodes, and the second content node is the previous content node of the first content node. The second content node and the first content node may belong to the same content group or different content groups, and this embodiment does not limit this.

[0058] Since in the scheme of this embodiment, when it is necessary to respond to the user's interactive behavior, the virtual digital object is driven to respond by inserting a response node, therefore, the presence of a response node between the second content node and the first content node can indicate that the virtual digital object has been driven between the second content node and the first content node to respond to the user's behavior.

[0059] like Figure 3 As shown, there are response nodes A1 and A2 between the second content node and the first content node, so the starting time of the current response interval is the starting time t2 of the response node A2.

[0060] By adopting the above scheme, the time range that drives the virtual digital object to interact with the user before the most recent content node can be included in the current response interval, which can more comprehensively and accurately determine the response object (for example, the question that needs to be answered, or the interactive behavior that needs to be responded to, etc.), which is conducive to optimizing the interactive effect of the virtual digital object.

[0061] In a non-limiting example, the start time of the current response interval may also depend on the type of the current responding node.

[0062] In the specific implementation, the B-side user can pre-set multiple types of interactive behaviors. The types of interactive behaviors correspond to the types of response nodes one by one, and the response nodes can be used for interactive behaviors that match the response types. The B-side user can set the number of response nodes of each type, and can also set the insertion order of various types of response nodes between two adjacent content nodes.

[0063] Corresponding to the types of interactive behaviors described above, the type of response node can be a question-and-answer type, or an interactive type. Among them, the question-and-answer type node can be used to answer questions raised by users, and the interactive type node can be used to respond to the user's interactive behavior. The B-end user can also set the interactive type response node to be inserted first, and then the question-and-answer type response node. Alternatively, the interactive type and question-and-answer type response nodes can also be inserted alternately. For ease of description, the question-and-answer type response nodes are referred to as question-and-answer nodes, and the interactive type response nodes are referred to as interactive nodes.

[0064] The starting time of the current response interval depends on the type of the current response node, which may mean: if there is a response node of the same type between the first content node and the second content node, the starting time of the current response interval is the starting time of the previous response node of the same type of the first content node, and the second content node is the previous content node of the first content node; if there is no response node of the same type between the first content node and the second content node, the starting time of the current response interval is the starting time of the first content node. The response node of the same type is a response node of the same type as the current response node.

[0065] If the response type corresponding to the current response interval is question reply, that is, the current response node is a question-answer node, and if there is a question-answer node between the second content node and the first content node, the start time of the current response interval is the start time of the previous question-answer node of the first content node. If there is no question-answer node between the second content node and the first content node, the start time of the current response interval is the start time of the first content node.

[0066] If the response type corresponding to the current response interval is an interactive response, that is, the current response node is an interactive node, and if there is an interactive node between the second content node and the first content node, the start time of the current response interval is the start time of the previous interactive node of the first content node. If there is no interactive node between the second content node and the first content node, the start time of the current response interval is the start time of the first content node.

[0067] like Figure 3As shown, assuming that the current response node is response node A3, and the type of response node A3 is question-and-answer type, if response node A1 and response node A2 are both question-and-answer nodes, then the starting time of the current response interval is the starting time t2 of response node A2. If response node A1 is a question-and-answer node, and response node A2 is an interactive node, then the starting time of the current response interval is the starting time t5 of response node A1. If response node A1 and response node A2 are not question-and-answer nodes, then the starting time of the current response interval is the starting time t1 of the first content node.

[0068] It should be noted that in the solution of this embodiment, the starting time of the response interval corresponding to each response node between two adjacent content nodes is the same. For example, when the current response node is response node A3, the starting time of the current response time interval is the same as when the current response node is response node A4.

[0069] The second aspect is about the end time of the current response interval.

[0070] In the solution of this embodiment, the end time of the current response interval is the end time of the previous node. In this embodiment, the type of the previous node is not limited. In this embodiment, the previous node can be the first content node. Or, the previous node can be the previous response node.

[0071] Specifically, if the current response node is the first response node after the first content node, the end time of the current response interval is the end time of the first content node. If the current response node is a non-first response node after the first content node, the end time of the current response interval is the end time of the previous response node of the current response node.

[0072] like Figure 3 As shown, if the current response node is response node A3, the end time of the current response interval is the end time t3 of the first content node. If the current response node is response node A4, the end time of the current response interval is the end time t4 of response node A3.

[0073] It should be noted that the end time of the current response interval has nothing to do with the type of response node. And the end time of the response interval corresponding to each response node between two adjacent content nodes is different. With the above scheme, the end time of the response time interval can be updated, and the time range for driving the virtual digital object to interact with the user after the most recent content node can be included in the current response interval, which can more comprehensively and accurately determine the response object (for example, the question that needs to be answered, or the interactive behavior that needs to be responded to, etc.), which is conducive to optimizing the interactive effect of the virtual digital object.

[0074] Furthermore, it is possible to determine whether to insert the current response node based on the user behavior information within the response time interval. If it is determined to be inserted, the response content of the current response node can be determined at the same time.

[0075] The solutions of the embodiments of the present application may include the following two situations, but are not limited thereto:

[0076] Case 1: The current response node is a question-and-answer node.

[0077] In this case, the question to be answered may be selected from a set of candidate questions, wherein the set of candidate questions includes standard questions that are hit within a current response interval, and the question to be answered has an unanswered label.

[0078] In a specific implementation, the B-end user can pre-set a set of standard questions, which may include multiple predefined standard questions, each with a pre-configured answer. During the live broadcast, whenever a question raised by a user is obtained, the NLP platform determines the hit standard question, and then can send the hit standard question to the B-end user's terminal. The B-end user's terminal can add the hit standard question to the candidate question set. If the candidate question set already has the hit standard question, the hit count of the standard question is updated. Thus, after the end of the current response interval, the question to be answered can be selected from the candidate question set.

[0079] It should be noted that after the end time of the current response interval, the number of times each standard question in the candidate question set is hit is the number of times the standard question is hit in the current response interval.

[0080] After the end of the current response interval, a question to be answered may be selected from the standard questions with an unanswered label in the candidate question set. In a specific implementation, having an unanswered label may mean being marked as unanswered or not marked as answered, and this embodiment does not limit this. The answered label in the candidate question set is marked after the first content node and before the current response node.

[0081] In a specific implementation, there is a one-to-one correspondence between the question-and-answer nodes and the questions to be answered, that is, each question-and-answer node can be used to drive the virtual digital object to answer a standard question.

[0082] For the current response node, a standard question can be randomly selected from the standard questions with unanswered labels in the candidate question set as the question to be answered, or the standard question with the most hits and unanswered labels can be selected from the candidate question set. Alternatively, when there are multiple standard questions with the most hits and unanswered labels, one can be randomly selected as the question to be answered.

[0083] Furthermore, the answer to the question to be answered may be read and used as the response content.

[0084] Furthermore, after determining the unanswered questions, the unanswered questions can be marked as answered or the unanswered labels can be removed.

[0085] It should be noted that if the current response node is the first question-answer node after the first content node, the label of each standard question in the candidate question set is unanswered. If the current response node is the last question-answer node between the first content node and the next content node, the candidate question set can also be cleared after the response content is determined.

[0086] It can be seen that in the solution of this embodiment, the label of answered or unanswered for the standard question is only valid between two adjacent content nodes. This solution can avoid repeated answers in a short period of time and avoid the situation where a question is not answered for a long time, which is conducive to optimizing the interaction effect between the virtual digital object and the user. For example, Figure 3 The response node A1 and the response node A2 are not used to reply to the same standard question, and the standard question replied by the response node A1 may be replied by the response node A3 or the response node A4.

[0087] It should be noted that if the number of times that the standard questions with unanswered labels in the candidate question set are hit exceeds a preset threshold, it can be determined not to insert the current response node.

[0088] In other embodiments, after determining the current response interval, the number of times each standard question in the current response interval is hit may be counted, and then questions to be answered may be selected from standard questions that have not been answered. Specifically, after determining the current response interval, the terminal of the B-end user may obtain questions raised by users in the current response interval, and the terminal of the B-end user may count the number of times each standard question in the current response interval is hit based on the questions raised by users in the current response interval, and then select questions to be answered from standard questions that have not been answered based on the number of hits.

[0089] Case 2: The current response node is an interactive node.

[0090] In this case, the highest priority and / or latest interactive operation can be determined within the current response interval and recorded as the pending response operation; then, the response content of the current response node is determined according to the pending response operation.

[0091] Specifically, the user behavior information within the current response interval can be read. The user behavior information within the current response interval can include multiple interactive behavior data. Each interactive behavior data can record the type of interactive behavior, the user ID that issued the interactive behavior, and the time when the user issued the interactive behavior.

[0092] In the first specific example, one of the multiple interactive behavior data in the current response interval can be randomly selected, and the interactive behavior recorded in the data is used as the behavior to be responded to.

[0093] In the second specific example, the latest interactive behavior of multiple interactive behavior data records within the current response interval can be used as the behavior to be responded to.

[0094] In the third specific example, multiple types of interactive behaviors may be preset, including but not limited to: sharing a live broadcast room, following a live broadcast room, entering a live broadcast room, and liking. The types of interactive behaviors may have a preset priority. Further, the interactive behavior with the highest priority from multiple interactive behavior data records may be used as the behavior to be responded to. Alternatively, the interactive behavior with the highest priority and the latest one may be used as the behavior to be responded to.

[0095] Furthermore, determining the response content according to the behavior to be responded to may include: generating the response content according to a preset response template. The response template may correspond to the type of interactive behavior one by one. For example, if the behavior to be responded to is a user entering the live broadcast room, the response content may be: "Welcome new baby, feel free to consult me ​​if you have any questions." In other embodiments, the response content may also be generated according to the user ID that issued the behavior to be responded to and the response template.

[0096] As an example, the order of priority from high to low for multiple interactive behaviors is: sharing the live broadcast room, following the live broadcast room, and entering the live broadcast room. If the user behavior information within the response time interval of the current response node records the interactive behavior of sharing the live broadcast room, you can respond to the most recent interactive behavior of sharing the live broadcast room. If the user behavior information does not record the interactive behavior of sharing the live broadcast room, you can respond to the most recent interactive behavior of following the live broadcast room. If the user behavior information does not record the interactive behavior of following the live broadcast room, you can respond to the most recent interactive behavior of entering the live broadcast room. If the above three interactive behaviors are not recorded, the current response node may not be inserted.

[0097] From the above, we can get the response content of the current response node.

[0098] In the specific implementation of step S23, the virtual digital object may be driven according to the response content.

[0099] Specifically, the virtual digital object can be driven according to the response content corresponding to the current response node, and multimodal information for interaction can be generated, and the multimodal information for interaction can be sent to the streaming server. The C-end user obtains the multimodal information for interaction from the streaming server, thereby realizing dynamic interaction between the virtual digital object and the user.

[0100] As a variation, a response node may be pre-configured between the second content node and the first content node, and the response node may be used to respond to the user's interactive behavior. The number of response nodes between two adjacent content nodes may be pre-set. More specifically, the B-side user may pre-set a response threshold, and the number of response nodes between two adjacent content nodes may be equal to the response threshold.

[0101] In a specific implementation, each response node can be used as the current response node in turn, and the response interval corresponding to the current response node can be determined and recorded as the current response interval. After determining the current response interval, the user behavior information in the current response interval can be counted to determine whether the current response node is used for response. If the current response node is not used for response, the current response node can be skipped. If it is determined that the current response node is used for response, the response content of the current response node can be determined based on the statistical results of the user behavior information in the current response interval.

[0102] In the first example, the start time of the first content node may be used as the start time of the current response interval.

[0103] In the second example, it may be first determined whether the response node between the first content node and the second content node is ignored, and the start time of the current response interval may be determined according to the determination result.

[0104] Specifically, a response node is also provided between the second content node and the first content node, and it can be determined whether the response node between the second content node and the first content node drives the virtual digital object to interact with the user. More specifically, it can be determined whether the response nodes between the second content node and the first content node are all ignored. If the judgment result is yes, the start time of the first content node can be used as the start time of the current response interval; otherwise, the start time of the previous non-ignored (that is, used for response) response node of the first content node can be used as the start time of the current response interval. It should be noted that "the response node is ignored" can mean that the response node is skipped, that is, the virtual digital object is not driven to interact with the user at the response node. In other words, the response node is not used for response.

[0105] In the third example, it is possible to first determine whether all response nodes of the same type between the first content node and the second content node are ignored, and determine the start time of the current response interval based on the determination result. If not all response nodes of the same type between the first content node and the second content node are ignored, the start time of the current response interval is the start time of the previous response node of the same type as the first content node. If all response nodes of the same type between the first content node and the second content node are ignored, the start time of the current response interval is the start time of the first content node.

[0106] In addition, if the current response node is the first response node among N response nodes between two adjacent content nodes, the end time of the current response interval is the end time of the first content node; if the current response node is the nth of the N response nodes, the end time of the current response interval is the end time of the nearest previous response node used for response before the nth response node, where 1 < n ≤ N and n is an integer.

[0107] Referring to Figure 4 , Figure 4 is a schematic flowchart of another live method for virtual digital objects in an embodiment of the present application. Figure 4 The method shown may include:

[0108] Step S41: When driving the virtual digital object to explain to the third content node, determine whether to insert a response group after the third content node;

[0109] Step S42: If the determination result is yes, after the third content node is explained, drive the virtual digital object to explain the response group.

[0110] Referring to Figure 5 , Figure 5 is a schematic diagram of the relationship between content groups in an embodiment of the present application.

[0111] The following combines Figure 5 to Figure 4 describe the method shown in a non-limiting manner.

[0112] In the specific implementation of step S41, the third content node may be the last content node in any content group in the main process. For example, the third content node may be Figure 5 content node 5 in content group 2 in

[0113] In specific implementation, step S41 may be executed during the process of driving the virtual digital object to explain the content corresponding to the third content node.

[0114] Specifically, it can be determined first based on the user behavior information within the group response interval whether to insert a response group and the response group to be inserted if it is determined to be inserted.

[0115] More specifically, the start time of the group response interval is the start time of the fourth content node, and the end time of the group response interval is the start time of the third content node. The fourth content node is the last content node in the previous content group of the current content group. Figure 5 As shown, the current content group is content group 2, the third content node is content node 5, and the fourth content node is the last content node of content group 1 (ie, content node 2).

[0116] Furthermore, whether to insert the response group can be determined according to the number of hits of each standard sentence in the group response interval. The standard sentence can be a declarative sentence or a question sentence, which is not limited in this embodiment.

[0117] Specifically, a standard sentence set may be pre-set, and the standard sentence set may include multiple predefined standard sentences, and each standard sentence has a corresponding response group. The response group in this embodiment also includes multiple content nodes, and each content node may have corresponding content. More specifically, the response group may be any one of the multiple content groups of the main process, or may be other groups other than the multiple content groups of the main process.

[0118] In a specific example, if at least one standard sentence is hit more than or equal to the group preset threshold in the group response interval, it can be determined to insert the response group. If the number of hits of each standard sentence in the group response interval is less than the group preset threshold, the response group may not be inserted after the third content node.

[0119] It should be noted that the number of times each standard sentence in the group response interval is hit can be counted after the group response interval is determined, or it can be counted in real time during the live broadcast. When the group response interval is determined, the number of times each standard sentence in the group response interval is hit can also be determined.

[0120] The standard sentence set and the standard question set can be different sets for different types of responses. In a specific implementation, after the NLP platform obtains the user behavior information, it can determine whether the hit is a standard sentence or a standard question, and then the B-side user's terminal updates the hit count accordingly.

[0121] In other embodiments, the standard sentence set and the standard question set belong to the same set. Specifically, a standard set may be pre-set, and the standard set may include multiple preset sentences, wherein each preset sentence may have a category label, and the category label of the preset sentence may be a standard sentence label or a standard question label. The preset sentences with standard sentence labels may be regarded as the above-mentioned standard sentences, and the preset sentences with standard question labels may be regarded as the above-mentioned standard question set. If the category label of the hit preset sentence is a standard sentence label, it can be determined that the hit preset sentence is a standard sentence. If the category label of the hit preset sentence is a standard question label, it can be determined that the hit preset sentence is a standard question.

[0122] Furthermore, in the case of determining to insert a response group, the response group corresponding to the standard sentence with the most hits in the group response interval can be used as the response group to be inserted. Alternatively, the response group corresponding to one of the standard sentences with the hit times greater than or equal to the group preset threshold in the group response interval can be randomly selected as the response group to be inserted. Alternatively, the response group corresponding to the standard sentence with the most hits in the group response interval can be used as the response group to be inserted.

[0123] From the above, after the current content group is explained, the virtual digital object can be driven to explain the content corresponding to the response group.

[0124] For example, the hit standard sentence may be "Explain product A", and the response group may be the product content group corresponding to product A. The product content group corresponding to product A may be the content group that has been explained before.

[0125] In a specific implementation, the priority of the response content may be higher than the response group.

[0126] Specifically, after the third content node is explained, the user behavior information in the current response interval is responded to first, and after the virtual digital object is driven according to the response content, the virtual digital object is driven to live broadcast the explanation response group. For the specific content of responding to the user behavior information in the current response interval, please refer to the above Figure 2 The relevant description will not be repeated here.

[0127] like Figure 5 As shown, after the virtual digital object is driven to explain the content node 5, if it is determined to insert the response group, the virtual digital object can be driven to explain the response group after the content node 5 is explained.

[0128] Or, if according to Figure 2The method shown determines that the response node A5 after the content node 5 needs to respond to the user behavior, so the virtual digital object can be driven to interact with the user at the response node A5, such as answering questions raised by the user or responding to the user's interactive behavior. After the response is completed (that is, after the response node A5), the virtual digital object is driven to explain the response group.

[0129] Further, after the virtual digital object is driven to explain the response group, the virtual digital object can continue to be driven along the main line process to explain. In other words, the virtual digital object can be driven to explain the next content group in the main line process according to the group sequence.

[0130] As can be seen from the above, in the solution of this embodiment, the user's need to explain a certain content group can be responded to between two adjacent content groups, thereby dynamically scheduling content groups for explanation according to the user's needs. Figure 2 In the described scheme, the start time and end time of the group response interval are both the start time of the content node. Compared with taking the end time of the last content node of two adjacent content groups as the start time and end time of the group response interval respectively, the above scheme takes into account that the scheduling time required for the response group is longer. Determining the response group in advance is conducive to reducing the virtual digital object explanation and making it more coherent and natural.

[0131] Based on the live broadcast method of virtual digital objects described above, the live broadcast process of the B-side user is described below in a non-restrictive manner.

[0132] B-side users can set up multiple content groups, content nodes of each content group and their corresponding content, group order, content node order, etc. in the live text library. They can also set up commonly used texts, which can refer to transition words. B-side users can also set up other groups, which can be explained in the branch process or in the group response process. B-side users can also set up Q&A content and interactive content in the live text library. Q&A content can include: standard question sets and answer content, etc. Interactive content can include interactive content response templates corresponding to various interactive behaviors. B-side users can add content, edit, delete, and other operations on content groups, commonly used texts, Q&A content, and interactive content.

[0133] B-side users can first create a live broadcast room and enter the live broadcast configuration page. B-side users can configure the contents described below on the live broadcast configuration page.

[0134] B-side users can set up virtual digital objects, including but not limited to: configuring the virtual digital objects' image, native language, timbre, posture, clothing, scene, illustrations, props, lens, resolution, etc.

[0135] B-side users can select already configured content groups and Q&A content in the live text library and add them to the settings list. New content groups, new Q&A content, and new interactive content can be added to the settings list. For content groups, Q&A content, and interactive content added to the settings list, you can add content, edit, delete, and other operations. For example, B-side users can select interactive content from the Q&A content that has been configured in the live text library and add it to the Q&A settings list. When editing, you can modify the reply type. If you modify or add a question, you need to retrain NLP to take effect.

[0136] For the multiple content nodes in the setting list above, the B-side user can also perform one or more of the following operations:

[0137] Operation 1: Configure the mall behavior, for example, display coupons, display product links, etc.

[0138] Operation 2: Insert common text. Common text can be inserted before and / or after the content corresponding to the content node.

[0139] Operation three: Delete any one or more content groups and / or content nodes and / or at least a portion of the content corresponding to the content nodes in the main process.

[0140] Operation 4: Automatically generate actions. If the B-side user selects at least one content group, actions and performance animations will be automatically generated based on the content corresponding to all nodes in the selected content group.

[0141] Operation 5: Edit the pronunciation content. Specifically, on the visual editing page, select the pronunciation content to be edited. The editing of the pronunciation content can include phonetic notation, pitch, speaking speed, volume, pauses, etc. The pronunciation content can be either explanation content or response content.

[0142] Operation 6: Manually generate an action. Specifically, drag the action to be added into the action timeline, that is, you can manually insert the action. For example, you can select an action in the action library.

[0143] Operation 7: Insert display information. Specifically, you can add display controls that need to appear in the display information timeline. Display controls can be used to display pictures, text, etc.

[0144] Operation 8: Add emoticons. Specifically, add the emoticons to be added to the emoticon timeline, that is, you can manually insert emoticons. For example, you can select emoticons in the emoticon library.

[0145] Except for operation 2, any other operations can be performed on question and answer content and interactive content.

[0146] In the live broadcast configuration page, one or more content groups are selected from the explanation list, and the order of the content groups is adjusted as the main process, that is, the virtual digital object is driven according to the content corresponding to the selected content group to generate multimodal information of the virtual digital object.

[0147] The B-side user can also set the number of Q&A response nodes and the number of interactive response nodes between two adjacent content nodes in the live broadcast configuration interface, for example, the maximum number of questions answered per interaction and the maximum number of responses per interaction. For example, if the number of Q&A response nodes and interactive response nodes are both 3, the order of the response nodes can be: interaction-Q&A-interaction-Q&A-interaction-Q&A. For another example, if the number of Q&A response nodes is 1 and the number of interactive response nodes is 3, the order of the response nodes can be: interaction-Q&A-interaction-interaction.

[0148] B-side users can also add different subtitle formats, previews, and preview clarity settings to virtual digital objects.

[0149] B-side users can also configure basic parameters during the live broadcast in the live broadcast configuration interface: triggering product card delay and group reply threshold.

[0150] Furthermore, after the B-side user configures the live broadcast content, open the live broadcast client and select the live broadcast room to be broadcast. Click the [Start] control to start streaming. After the streaming is successfully pushed, the captured image is successfully pushed to the live broadcast room.

[0151] From the above, the solution provided by the embodiment of the present application has the following beneficial effects:

[0152] 1. Full-stack, end-to-end solution

[0153] The solution of this embodiment can be connected to and adapt to various product forms, such as APP, mini-programs, and web pages; it can also support cloud deployment and local deployment to meet various live broadcast application scenarios.

[0154] 2. Multiple types of high-quality avatars

[0155] It can support a variety of virtual image options and also supports customization, including hyper-realism, three-dimensional beauty, cartoon, two-dimensional and other types, which can cover the company's personal IP, brand IP and product IP needs.

[0156] 3. Real-time intelligent synthesis of high-quality voice animation

[0157] Through text to speech and animation technology (TTSA), speech to animation technology (STA), expressive text to speech technology (ETTS), etc., it is possible to generate high-quality speech and animation of virtual humans based on text and speech in real time, making virtual humans as vivid and lifelike as real people.

[0158] 4. Natural and smooth movements and expressions are intelligently generated

[0159] Through the intelligent generation technology of action and expression, virtual humans can be driven to generate smooth body movements and natural facial expressions.

[0160] 5. Efficient delivery of industrialized intelligent production lines

[0161] AI-driven, intelligent modeling, digital collaboration, and automated processes are used to solve collaboration and customer demand management issues.

[0162] 6. Multimodal AI intelligent explanation: Based on the script content and virtual human presentation in the live broadcast room, multimodal intelligent explanation of virtual humans is realized.

[0163] 7. Multimodal AI intelligent interaction: Based on the intelligent dialogue system and virtual human presentation, multimodal intelligent interaction between users and virtual humans can be realized.

[0164] Reference Figure 6 , Figure 6 is a structural diagram of a live broadcast device for a virtual digital object in an embodiment of the present application. Figure 6 The apparatus shown may include:

[0165] The acquisition module 61 is used to acquire user behavior information in the process of driving the live broadcast of the virtual digital object according to the explanation content, wherein the explanation content includes a plurality of content nodes;

[0166] A response determination module 62, configured to determine the response content of the current response node according to the user behavior information in the current response interval after the first content node is explained, wherein the first content node is any one of the plurality of content nodes;

[0167] The driving module 63 is used to drive the live broadcast of the virtual digital object according to the response content.

[0168] For more information about the working principle, working method, beneficial effects, etc. of the live broadcast device for virtual digital objects in the embodiments of the present application, please refer to the above description of the live broadcast method for virtual digital objects, which will not be repeated here.

[0169] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned live broadcast method of the virtual digital object are executed. The storage medium may include ROM, RAM, a magnetic disk or an optical disk, etc. The storage medium may also include a non-volatile memory (non-volatile) or a non-transitory memory, etc.

[0170] The embodiment of the present application also provides a terminal, including a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor executes the steps of the above-mentioned method for live broadcasting of virtual digital objects when running the computer program. The terminal includes but is not limited to terminal devices such as mobile phones, computers, and tablet computers.

[0171] It should be understood that in the embodiments of the present application, the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0172] It should also be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM) and direct rambus RAM (DR RAM).

[0173] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer program may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless means.

[0174] In the several embodiments provided in the present application, it should be understood that the disclosed methods, devices and systems can be implemented in other ways. For example, the device embodiments described above are merely schematic; for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation; for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0175] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may be physically included separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units. For example, for each device or product applied to or integrated in a chip, each module / unit contained therein may be implemented in the form of hardware such as circuits, or at least some of the modules / units may be implemented in the form of software programs, which run on a processor integrated inside the chip, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits; for each device or product applied to or integrated in a chip module, each module / unit contained therein may be implemented in the form of hardware such as circuits, and different modules / units may be located in the same component (such as a chip, circuit module, etc.) or different components of the chip module, or at least some of the modules / units may be implemented in the form of software programs. The software programs may run on a processor integrated inside the chip, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits. The element can be implemented in the form of a software program, which runs on a processor integrated inside the chip module, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits; for various devices and products applied to or integrated in the terminal, the various modules / units contained therein can be implemented in the form of hardware such as circuits, and different modules / units can be located in the same component (for example, chip, circuit module, etc.) or in different components in the terminal, or, at least some modules / units can be implemented in the form of a software program, which runs on a processor integrated inside the terminal, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits.

[0176] It should be understood that the term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article indicates that the associated objects before and after are in an "or" relationship.

[0177] The "multiple" appearing in the embodiments of the present application refers to two or more. The first, second, etc. descriptions appearing in the embodiments of the present application are only used for illustration and distinction of the description objects. There is no order, nor does it represent a special limitation on the number of devices in the embodiments of the present application, and cannot constitute any limitation on the embodiments of the present application. Although the present application is disclosed as above, the present application is not limited to this. Any person skilled in the art may make various changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims.

[0178] Although the present application is disclosed as above, the present application is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application. Therefore, the protection scope of the present application shall be subject to the scope defined by the claims.

Claims

1. A method for live broadcasting of a virtual digital object, characterized in that: The method comprises: In the process of driving the live broadcast of the virtual digital object according to the explanation content, obtaining user behavior information, wherein the explanation content includes a plurality of content nodes; After the first content node is explained, the response content of the current response node is determined according to the user behavior information in the current response interval, wherein the first content node is any one of the multiple content nodes; wherein the current response node is the first or non-first response node after the first content node; the start time and end time of the current response interval are dynamically adjusted for the current response node; wherein the start time of the current response interval is the start time of the first content node, or the start time of a response node before the first content node; the end time of the current response interval is the end time of the first content node, or the end time of a response node before the current response node; The virtual digital object is driven to be broadcast live according to the response content.

2. The live broadcast method of virtual digital objects according to claim 1, characterized in that: If there is a response node between the first content node and the second content node, the start time of the current response interval is the start time of the previous response node of the first content node, and the second content node is the previous content node of the first content node; If there is no response node between the first content node and the second content node, the start time of the current response interval is the start time of the first content node.

3. The live broadcast method of virtual digital objects according to claim 1, characterized in that: If there is a response node of the same type between the first content node and the second content node, the start time of the current response interval is the start time of the previous response node of the same type of the first content node, and the second content node is the previous content node of the first content node; If there is no response node of the same type between the first content node and the second content node, the start time of the current response interval is the start time of the first content node; The same type of response node is a response node of the same type as the current response node.

4. The live broadcast method of virtual digital objects according to claim 1, characterized in that: The current response node is an interactive node, and the interactive node is used to respond to the user's interactive behavior. Determining the response content of the current response node includes: Determine the highest priority and / or latest interactive behavior in the current response interval and record it as the behavior to be responded; According to the behavior to be responded, the response content of the current responding node is determined.

5. The live broadcast method of virtual digital objects according to claim 1, characterized in that: The current response node is a question-and-answer node, which is used to answer questions raised by users. Determining the response content of the current response node includes: Selecting a question to be answered from a set of candidate questions, wherein the set of candidate questions includes standard questions that are hit within the current response interval, and the question to be answered has an unanswered label; Determine the answer to the question to be answered.

6. The live broadcast method of virtual digital objects according to claim 5, characterized in that: If the current response node is the first question-and-answer node after the first content node, each standard question in the candidate question set has the unanswered label.

7. The live broadcast method of virtual digital objects according to claim 1, characterized in that: The explanation content includes a plurality of content groups, each content group includes at least one content node, and the method further includes: When the virtual digital object is driven to explain to the third content node, it is determined whether a response group is inserted after the third content node, the response group being one of the plurality of content groups or other group other than the plurality of content groups, and the third content node being the last content node in any one of the content groups; If the judgment result is yes, after the third content node is explained, the virtual digital object is driven to explain the response group.

8. The live broadcast method of virtual digital objects according to claim 7, characterized in that: Determining whether to insert a response group after the third content node includes: Determine the hit standard sentence in the group response interval, wherein the start time of the group response interval is the start time of the fourth content node, the end time is the start time of the third content node, and the fourth content node is the last content node in the previous content group of the current content group; It is determined whether there is a standard sentence among the hit standard sentences whose hit times is greater than a preset threshold of the group. If yes, insert it into the response group.

9. A live broadcast device for a virtual digital object, characterized in that: The device comprises: An acquisition module, used to acquire user behavior information in the process of driving the live broadcast of the virtual digital object according to the explanation content, wherein the explanation content includes a plurality of content nodes; A response determination module, used to determine the response content of the current response node according to the user behavior information in the current response interval after the first content node has been explained, wherein the first content node is any one of the multiple content nodes; wherein the current response node is the first or non-first response node after the first content node; the start time and end time of the current response interval are dynamically adjusted for the current response node; wherein the start time of the current response interval is the start time of the first content node, or the start time of a response node before the first content node; and the end time of the current response interval is the end time of the first content node, or the end time of a response node before the current response node; A driving module is used to drive the live broadcast of the virtual digital object according to the response content.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for live broadcasting of a virtual digital object as claimed in any one of claims 1 to 8 are executed.

11. A terminal comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, characterized in that: When the processor runs the computer program, the steps of the method for live broadcasting of a virtual digital object described in any one of claims 1 to 8 are executed.

Images