A cross-platform multi-screen interactive method

Through real-time three-dimensional models and artificial intelligence models, the problem of cumbersome interaction methods and inability to respond instantly in multi-screen interactive technology is solved, real-time interaction is achieved for users, improving the experience and preventing inappropriate participation.

CN117492688BActive Publication Date: 2024-11-19BEIJING REDIO CULTURE COMM CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311664853.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-06
Publication Date
2024-11-19
Estimated Expiration
2043-12-06

AI Technical Summary

Technical Problem

The existing multi-screen interactive technology has cumbersome operations and inability to respond instantly in real-time interaction, resulting in poor user experience.

Method used

By establishing a real-time three-dimensional model, combining artificial intelligence models, obtaining and integrating information from the active terminal and interactive terminal, generating active simulation audio and setting audio thresholds, judging the participation conditions of the interactive terminal, and real-time interactive judgment is achieved.

Benefits of technology

It enables users to interact in real time based on the event content, improves the viewer's experience, prevents the participation of absentees and noise interference, and ensures the accuracy and security of the interaction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117492688B_ABST
    Figure CN117492688B_ABST
Patent Text Reader

Abstract

The present application discloses a cross-platform multi-screen interaction method, which relates to the technical field of multimedia data playback, and solves the problem that the interaction method in the existing multi-screen interaction is too cumbersome and cannot be used for real-time interaction, resulting in a poor viewing experience for participants; step one: obtaining the activity terminal information of the activity terminal and the interaction terminal information of the interaction terminal; step two: generating a real-time three-dimensional model; step three: obtaining the activity simulation audio through the simulation model; step four: obtaining the audio threshold corresponding to the interaction terminal through the threshold generation model; step five: setting participation condition one according to the audio similarity obtained based on the activity simulation audio and the corresponding interaction collection audio; generating participation condition two according to the interaction collection action; step six: judging whether the interaction terminal participates in the activity based on participation condition one and participation condition two; the present application determines whether the interaction terminal participates in the activity by judging the real-time interaction situation of the interaction terminal.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the field of multimedia data playback technology, and relates to multimedia data playback technology, and specifically to a cross-platform multi-screen interactive method. Background Art

[0002] With the development of science and technology, multi-screen interaction has a wide range of application scenarios, such as home entertainment, business office, education and learning, and performance activities. Realizing seamless connection and information sharing between different devices brings a lot of convenience to our life and work.

[0003] Existing multi-screen interactive technology When we watch video programs or participate in large offline activities, using a large screen or smart TV is actually a very important scenario. At this time, most of the interactions are conducted online or by superimposing QR codes on the screen. However, in practice, these two methods are not very good choices. The first method is actually suitable for single-screen interaction. The second method requires users to stop the current state and open their mobile phones to scan the code, which seriously affects the user experience. Moreover, when the content of the activity changes, the user's interaction cannot be changed immediately, that is, personalized interaction cannot be performed at any time according to the content, which seriously affects the viewer's experience. Therefore, a cross-platform multi-screen interactive method is needed to enable viewers to interact in real time according to the content of the activity and have a better activity experience. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a cross-platform multi-screen interaction method to solve the problem that the interaction method in the existing multi-screen interaction is too cumbersome and cannot be used for real-time interaction, resulting in a poor viewing experience for participants; the present application solves the above-mentioned problem by obtaining the real-time status of the interactive terminal in real time to determine whether the interactive terminal meets the interaction requirements.

[0005] To achieve the above-mentioned purpose, the first aspect of the present application provides a cross-platform multi-screen interaction method, comprising the following steps:

[0006] Step 1: Obtain activity terminal information of an activity terminal and interactive terminal information of several interactive terminals; wherein the activity terminal information includes activity terminal location, activity content, and activity audio information; the interactive terminal information includes interactive terminal location, interactive collected audio, and interactive collected action;

[0007] Step 2: Building a three-dimensional model based on the active terminal position and several interactive terminal positions, and marking several real-time interactive terminal positions in the three-dimensional model to generate a real-time three-dimensional model;

[0008] Step 3: Identify and obtain the activity terminal location, activity audio information and interactive terminal location, and integrate them into input data, input into the simulation model, and obtain activity simulation audio; wherein the simulation model is obtained through artificial intelligence model training, and the activity audio information includes interactive keywords and their corresponding audio length and output volume;

[0009] Step 4: The active terminal information, interactive terminal information and real-time three-dimensional model are sorted and input into a threshold generation model to obtain an audio threshold corresponding to the interactive terminal; wherein the threshold generation model is obtained by training an artificial intelligence model;

[0010] Step 5: Based on the audio similarity obtained by processing the activity simulation audio and the corresponding interactive collection audio, setting the first participation condition based on the audio similarity; and determining whether the interactive terminal generates an interactive collection action, and generating the second participation condition; wherein the value of the first participation condition is "0" or "1"; and the value of the second participation condition is "0" or "1";

[0011] Step 6: Determine whether the interactive terminal participates in the activity based on participation condition 1 and participation condition 2.

[0012] The present application establishes a real-time three-dimensional model, and simulates the interactive collection audio of the interactive terminal through the real-time three-dimensional model to generate active simulation audio, and sets different audio thresholds for interactive terminals in different positions through the real-time three-dimensional model and the active terminal information and the interactive terminal information; and by comparing the relationship between the similarity of the interactive collection audio and the active simulation audio and the audio threshold, and whether the interactive terminal performs an interactive collection action, the real-time interaction situation of the interactive terminal is judged, and then it is judged whether the interactive terminal has successfully participated in the activity.

[0013] Preferably, the simulation model is obtained by training an artificial intelligence model, including:

[0014] Integrate the activity terminal positions and their corresponding activity audio information in several groups of historical data, as well as the interactive terminal positions and their corresponding interactive collected audio, into training data to train the artificial intelligence model; input the activity audio information, activity terminal positions and interactive terminal positions to obtain the activity simulation audio corresponding to the interactive terminal;

[0015] The artificial intelligence model is trained by using model input data and model output data to obtain a simulation model; wherein the artificial intelligence model includes a BP neural network model or a RBF neural network model.

[0016] The present application trains an artificial intelligence model through the activity terminal locations and their corresponding activity audio information in a large amount of historical data, as well as the interactive terminal locations and their corresponding interactive collected audio, and then obtains a simulation model, and generates activity simulation audio through the simulation model and the real-time activity terminal locations and interactive terminal locations, making the activity simulation audio more reliable.

[0017] Preferably, the obtaining of the audio threshold corresponding to the interactive terminal based on the active terminal information, the interactive terminal information and the real-time three-dimensional model includes:

[0018] Identify and obtain the audio length and output volume corresponding to the interactive keywords in the active terminal information, and integrate them into inherent factors; obtain the distance between the interactive terminal and the active terminal, and the number of other interactive terminals between the interactive terminal and the active terminal through a real-time three-dimensional model, and integrate them into real-time variable factors;

[0019] Integrate a number of inherent factors and real-time variable factors, the interactive audio collected by the corresponding interactive terminals, and the activity audio information into training data to train the artificial intelligence model;

[0020] Integrate several inherent factors and real-time variable factors, the interactive audio collected by the corresponding interactive terminal, and the activity audio information into experimental data, input them into the trained artificial intelligence model, and output the audio threshold; and adjust the parameters of the trained artificial intelligence model according to the output audio threshold, and finally obtain the threshold generation model;

[0021] The artificial intelligence model is trained by using model input data and model output data to obtain a threshold generation model; wherein the artificial intelligence model includes a BP neural network model or a RBF neural network model.

[0022] The present application obtains the audio length and output volume corresponding to the interactive keywords in a large amount of active terminal information by identifying and integrating them into inherent factors; obtains the distance between the interactive terminal and the active terminal, as well as the number of other interactive terminals between the interactive terminal and the active terminal through a real-time three-dimensional model, and integrates them into real-time variable factors; integrates several inherent factors and real-time variable factors and the interactive collected audio corresponding to the corresponding interactive terminal, as well as the active audio information into training data, taking into account the number of other interactive terminals between the interactive terminal and the active terminal, and training the artificial intelligence model to obtain a threshold generation model, and through the threshold generation model, it is possible to generate an audio threshold for the real-time interactive terminal position of the interactive terminal, so that the audio threshold is more adaptable and more reliable.

[0023] Preferably, the activity content includes interaction time, interaction actions and interaction rewards; wherein the interaction time is the activity duration set by the organizer, the interaction action is the action to check whether the interactive terminal meets the requirements of this activity, and the interaction reward is the reward given to the interactive terminal that completes this activity.

[0024] Preferably, the audio similarity obtained by processing the activity simulation audio and the corresponding interactive collection audio includes:

[0025] The interactive collected audio and the active simulated audio are sequentially preprocessed, feature extracted and audio signal aligned to obtain the comparative collected audio and comparative simulated audio; the comparative collected audio and the comparative simulated audio are similarly calculated to obtain the audio similarity; wherein the preprocessing includes noise reduction and removal of repeated sounds; the feature extraction includes extraction of timbre features, pitch features and rhythm features; and the audio signal alignment uses dynamic time warping to perform audio alignment.

[0026] The similarity calculation in this application can be implemented using existing technologies such as Euclidean distance and cosine similarity.

[0027] Preferably, the step of setting the first participation condition based on audio similarity includes:

[0028] Determine whether the audio similarity is greater than or equal to the audio threshold;

[0029] If yes, set the value of participation condition 1 to "1";

[0030] If not, determine whether the difference between the audio similarity of other interactive terminals closest to the interactive terminal and the audio similarity is less than or equal to the difference threshold; when the difference is less than or equal to the difference threshold, set the value of participation condition one to "1"; otherwise, set the value of participation condition one to "0".

[0031] The present application limits the audio similarity between the interactively collected audio and the activity simulation audio by setting an audio threshold, and sets the value of the participation condition one corresponding to the interactive terminal whose audio similarity is greater than the audio threshold to "1", so as to prevent absent persons from participating in the activity; at the same time, by comparing the audio similarities of other interactive terminals around the interactive terminal with the audio similarity of the interactive terminal to see whether they are less than the difference threshold, it is convenient to prevent the interactive terminal from being unable to participate in the activity due to the influence of excessive noise in some areas on the interactive collected audio of the interactive terminal.

[0032] Preferably, the step of judging whether the interactive terminal generates an interactive collection action and generating the second participation condition includes:

[0033] Determining whether the interactive terminal generates an interactive collection action;

[0034] If yes, set the value of participation condition 2 to "1";

[0035] If not, a confirmation message is sent to the corresponding interactive terminal, and it is determined whether the interactive terminal participates in the activity; when the interactive terminal participates in the activity, the corresponding interactive terminal is reminded to perform the interactive action, and the value of the second participation condition is set to "1"; otherwise, the value of the second participation condition is set to "0".

[0036] Preferably, the determining whether the interactive terminal participates in the activity based on the first participation condition and the second participation condition includes:

[0037] Determine whether the values ​​of participation condition 1 and participation condition 2 are both "1";

[0038] If yes, it is determined that the corresponding interactive terminal participates in the activity;

[0039] If not, it is determined that the corresponding interactive terminal does not participate in the activity.

[0040] This application combines participation condition one and participation condition two to jointly determine whether the interactive terminal can successfully participate in the activity, making the determination more accurate.

[0041] Compared with the prior art, the beneficial effects of this application are:

[0042] 1. This application establishes a real-time three-dimensional model, and simulates the interactive collection audio of the interactive terminal through the real-time three-dimensional model to generate active simulation audio, and sets different audio thresholds for interactive terminals in different positions through the real-time three-dimensional model and the active terminal information and interactive terminal information; and by comparing the relationship between the similarity of the interactive collection audio and the active simulation audio and the audio threshold, as well as whether the interactive terminal performs an interactive collection action, the real-time interaction situation of the interactive terminal is judged, and then it is judged whether the interactive terminal has successfully participated in the activity.

[0043] 2. The present application limits the audio similarity between the interactively collected audio and the activity simulation audio by setting an audio threshold, and sets the value of the participation condition one corresponding to the interactive terminal whose audio similarity is greater than the audio threshold to "1", so as to prevent absent persons from participating in the activity; at the same time, by comparing the audio similarities of other interactive terminals around the interactive terminal with the audio similarity of the interactive terminal to see whether they are less than the difference threshold, it is convenient to prevent the interactive terminal from being unable to participate in the activity due to the influence of excessive noise in some areas on the interactive collected audio of the interactive terminal. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 A flowchart of the steps of this application;

[0046] Figure 2 Set up a flow chart for condition 1 of this application;

[0047] Figure 3 Set up a flow chart for condition 2 of this application. DETAILED DESCRIPTION

[0048] The technical solution of the present application will be described clearly and completely in conjunction with the embodiments below. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0049] See also Figure 1-Figure 3 The first aspect of the present application provides a cross-platform multi-screen interaction method, including the following steps:

[0050] Step 1: Obtain activity terminal information of an activity terminal and interactive terminal information of several interactive terminals; wherein the activity terminal information includes activity terminal location, activity content, and activity audio information; the interactive terminal information includes interactive terminal location, interactive collected audio, and interactive collected action;

[0051] Step 2: Building a three-dimensional model based on the active terminal position and several interactive terminal positions, and marking several real-time interactive terminal positions in the three-dimensional model to generate a real-time three-dimensional model;

[0052] Step 3: Identify and obtain the activity terminal location, activity audio information and interactive terminal location, and integrate them into input data, input into the simulation model, and obtain activity simulation audio; wherein the simulation model is obtained through artificial intelligence model training, and the activity audio information includes interactive keywords and their corresponding audio length and output volume;

[0053] Step 4: The active terminal information, interactive terminal information and real-time three-dimensional model are sorted and input into a threshold generation model to obtain an audio threshold corresponding to the interactive terminal; wherein the threshold generation model is obtained by training an artificial intelligence model;

[0054] Step 5: Based on the audio similarity obtained by processing the activity simulation audio and the corresponding interactive collection audio, setting the first participation condition based on the audio similarity; and determining whether the interactive terminal generates an interactive collection action, and generating the second participation condition; wherein the value of the first participation condition is "0" or "1"; and the value of the second participation condition is "0" or "1";

[0055] Step 6: Determine whether the interactive terminal participates in the activity based on participation condition 1 and participation condition 2.

[0056] The present application establishes a real-time three-dimensional model, and simulates the interactive collection audio of the interactive terminal through the real-time three-dimensional model to generate active simulation audio, and sets different audio thresholds for interactive terminals in different positions through the real-time three-dimensional model and the active terminal information and the interactive terminal information; and by comparing the relationship between the similarity of the interactive collection audio and the active simulation audio and the audio threshold, and whether the interactive terminal performs an interactive collection action, the real-time interaction situation of the interactive terminal is judged, and then it is judged whether the interactive terminal has successfully participated in the activity.

[0057] The simulation model is obtained through artificial intelligence model training, including:

[0058] Integrate the activity terminal positions and their corresponding activity audio information in several groups of historical data, as well as the interactive terminal positions and their corresponding interactive collected audio, into training data to train the artificial intelligence model; input the activity audio information, activity terminal positions and interactive terminal positions to obtain the activity simulation audio corresponding to the interactive terminal;

[0059] The artificial intelligence model is trained by using model input data and model output data to obtain a simulation model; wherein the artificial intelligence model includes a BP neural network model or a RBF neural network model.

[0060] Through the activity terminal position in the activity audio information in the historical data, the interactive terminal position in the activity audio information and the interactive terminal information, and the interactive collected audio, these data are processed and then the artificial intelligence model is trained to obtain the activity terminal position, interactive keywords, audio length and output volume, as well as the relationship between the interactive terminal position and the interactive collected audio; the training data can be group training, which can be the same activity audio information and activity terminal position, but different interactive terminal positions; the same interactive terminal position and activity terminal position, but different activity audio information; group training of the artificial intelligence model; that is, the final input of the activity audio information, the activity terminal position and the interactive terminal position, the simulated interactive collected audio output is the activity simulated audio; group training can effectively obtain the simulation model.

[0061] The present application trains an artificial intelligence model through the activity terminal locations and their corresponding activity audio information in a large amount of historical data, as well as the interactive terminal locations and their corresponding interactive collected audio, and then obtains a simulation model, and generates activity simulation audio through the simulation model and the real-time activity terminal locations and interactive terminal locations, making the activity simulation audio more reliable.

[0062] The audio threshold corresponding to the interactive terminal is obtained based on the active terminal information, the interactive terminal information and the real-time three-dimensional model, including:

[0063] Identify and obtain the audio length and output volume corresponding to the interactive keywords in the active terminal information, and integrate them into inherent factors; obtain the distance between the interactive terminal and the active terminal, and the number of other interactive terminals between the interactive terminal and the active terminal through a real-time three-dimensional model, and integrate them into real-time variable factors;

[0064] Integrate a number of inherent factors and real-time variable factors, the interactive audio collected by the corresponding interactive terminals, and the activity audio information into training data to train the artificial intelligence model;

[0065] Integrate several inherent factors and real-time variable factors, the interactive audio collected by the corresponding interactive terminal, and the activity audio information into experimental data, input them into the trained artificial intelligence model, and output the audio threshold; and adjust the parameters of the trained artificial intelligence model according to the output audio threshold, and finally obtain the threshold generation model;

[0066] The artificial intelligence model is trained by using model input data and model output data to obtain a threshold generation model; wherein the artificial intelligence model includes a BP neural network model or a RBF neural network model.

[0067] The present application obtains the audio length and output volume corresponding to the interactive keywords in a large amount of active terminal information by identifying and integrating them into inherent factors; obtains the distance between the interactive terminal and the active terminal, as well as the number of other interactive terminals between the interactive terminal and the active terminal through a real-time three-dimensional model, and integrates them into real-time variable factors; integrates several inherent factors and real-time variable factors and the interactive collected audio corresponding to the corresponding interactive terminal, as well as the active audio information into training data, taking into account the number of other interactive terminals between the interactive terminal and the active terminal, and training the artificial intelligence model to obtain a threshold generation model, and through the threshold generation model, it is possible to generate an audio threshold for the real-time interactive terminal position of the interactive terminal, so that the audio threshold is more adaptable and more reliable.

[0068] The activity content includes interaction time, interaction actions and interaction rewards; the interaction time is the activity duration set by the organizer, the interaction action is the action to check whether the interactive terminal meets the requirements of this activity, and the interaction reward is the reward given to the interactive terminal that completes this activity; when the interactive terminal successfully participates in the corresponding activity, it can obtain the corresponding interaction reward.

[0069] The audio similarity obtained by processing the activity simulation audio and the corresponding interactive collection audio includes:

[0070] The interactive collected audio and the active simulated audio are sequentially preprocessed, feature extracted and audio signal aligned to obtain the comparative collected audio and comparative simulated audio; the comparative collected audio and the comparative simulated audio are similarly calculated to obtain the audio similarity; wherein the preprocessing includes noise reduction and removal of repeated sounds; the feature extraction includes extraction of timbre features, pitch features and rhythm features; and the audio signal alignment uses dynamic time warping to perform audio alignment.

[0071] The similarity calculation in this application can be implemented using existing technologies such as Euclidean distance and cosine similarity.

[0072] Participation condition 1 is set based on audio similarity, including:

[0073] Determine whether the audio similarity is greater than or equal to the audio threshold;

[0074] If yes, set the value of participation condition 1 to "1";

[0075] If not, determine whether the difference between the audio similarity of other interactive terminals closest to the interactive terminal and the audio similarity is less than or equal to the difference threshold; when the difference is less than or equal to the difference threshold, set the value of participation condition one to "1"; otherwise, set the value of participation condition one to "0".

[0076] The present application limits the audio similarity between the interactively collected audio and the activity simulation audio by setting an audio threshold, and sets the value of the participation condition one corresponding to the interactive terminal whose audio similarity is greater than the audio threshold to "1", so as to prevent absent persons from participating in the activity; at the same time, by comparing the audio similarities of other interactive terminals around the interactive terminal with the audio similarity of the interactive terminal to see whether they are less than the difference threshold, it is convenient to prevent the interactive terminal from being unable to participate in the activity due to the influence of excessive noise in some areas on the interactive collected audio of the interactive terminal.

[0077] Determine whether the interactive terminal generates an interactive collection action, and generate the second participation condition, including:

[0078] Determining whether the interactive terminal generates an interactive collection action;

[0079] If yes, set the value of participation condition 2 to "1";

[0080] If not, a confirmation message is sent through the interactive terminal, and it is determined whether the interactive terminal participates in the activity; when the interactive terminal participates in the activity, the corresponding interactive terminal is reminded to perform the interactive action, and the value of the second participation condition is set to "1"; otherwise, the value of the second participation condition is set to "0".

[0081] Determining whether the interactive terminal participates in the activity based on the first participation condition and the second participation condition includes:

[0082] Determine whether the values ​​of participation condition 1 and participation condition 2 are both "1";

[0083] If yes, it is determined that the interactive terminal has successfully participated in the activity;

[0084] If not, it is determined that the interactive terminal fails to participate in the activity.

[0085] When the value of the interactive terminal's participation condition 1 is "1"

[0086] This application combines participation condition one and participation condition two to jointly determine whether the interactive terminal can successfully participate in the activity, making the determination more accurate.

[0087] How this application works:

[0088] When used, the present application first obtains the activity terminal information of the activity terminal and the interactive terminal information of several interactive terminals; and establishes a three-dimensional model according to the activity terminal position and the several interactive terminal positions, and marks several real-time interactive terminal positions in the three-dimensional model to generate a real-time three-dimensional model; identifies and obtains the activity terminal position, activity audio information and interactive terminal position, and integrates them into input data, inputs the simulation model, and obtains the activity simulation audio; organizes the activity terminal information, the interactive terminal information and the real-time three-dimensional model and inputs them into the threshold generation model to obtain the audio threshold corresponding to the interactive terminal; obtains the audio similarity according to the activity simulation audio and the corresponding interactive collection audio, and sets the participation condition one according to the audio similarity; and, determines whether the interactive terminal generates an interactive collection action, and generates the participation condition two; determines whether the interactive terminal participates in the activity according to the participation condition one and the participation condition two; wherein the participation audio threshold, the activity simulation audio and the interactive collection action are all obtained in real time and change according to the real-time changes, which solves the problem that the user interaction in the existing cross-platform multi-screen interaction method cannot be changed at any time, that is, personalized interaction cannot be performed at any time according to the content, which affects the viewer's experience, that is, solves the technical problem that the existing multi-screen interaction method is difficult to achieve real-time interaction.

[0089] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical method of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A cross-platform multi-screen interactive method, characterized in that: The following steps are involved: Step 1: Obtain activity terminal information of an activity terminal and interactive terminal information of several interactive terminals; wherein the activity terminal information includes activity terminal location, activity content, and activity audio information; the interactive terminal information includes interactive terminal location, interactive collected audio, and interactive collected action; Step 2: Building a three-dimensional model based on the active terminal position and several interactive terminal positions, and marking several real-time interactive terminal positions in the three-dimensional model to generate a real-time three-dimensional model; Step 3: Identify and obtain the activity terminal location, activity audio information and interactive terminal location, and integrate them into input data, input into the simulation model, and obtain activity simulation audio; wherein the simulation model is obtained through artificial intelligence model training, and the activity audio information includes interactive keywords and their corresponding audio length and output volume; The simulation model is obtained through artificial intelligence model training, including: Integrate the activity terminal positions and their corresponding activity audio information in several groups of historical data, as well as the interactive terminal positions and their corresponding interactive collected audio, into training data to train the artificial intelligence model; input the activity audio information, activity terminal positions and interactive terminal positions to obtain the activity simulation audio corresponding to the interactive terminal; The artificial intelligence model is trained by using the model input data and the model output data to obtain a simulation model; wherein the artificial intelligence model includes a BP neural network model or a RBF neural network model; Step 4: The active terminal information, interactive terminal information and real-time three-dimensional model are sorted and input into a threshold generation model to obtain an audio threshold corresponding to the interactive terminal; wherein the threshold generation model is obtained by training an artificial intelligence model; The threshold generation model is obtained through artificial intelligence model training, including: Identify and obtain the audio length and output volume corresponding to the interactive keywords in the active terminal information, and integrate them into inherent factors; obtain the distance between the interactive terminal and the active terminal, and the number of other interactive terminals between the interactive terminal and the active terminal through a real-time three-dimensional model, and integrate them into real-time variable factors; Integrate a number of inherent factors and real-time variable factors, the interactive audio collected by the corresponding interactive terminals, and the activity audio information into training data to train the artificial intelligence model; Integrate several inherent factors and real-time variable factors, the interactive audio collected by the corresponding interactive terminal, and the activity audio information into experimental data, input them into the trained artificial intelligence model, and output the audio threshold; and adjust the parameters of the trained artificial intelligence model according to the output audio threshold, and finally obtain the threshold generation model; The artificial intelligence model is trained by using the model input data and the model output data to obtain a threshold generation model; wherein the artificial intelligence model includes a BP neural network model or a RBF neural network model; Step 5: Based on the audio similarity obtained by processing the activity simulation audio and the corresponding interactive collection audio, setting the first participation condition based on the audio similarity; and determining whether the interactive terminal generates an interactive collection action, and generating the second participation condition; The first step of setting the participation condition based on the audio similarity includes: Determine whether the audio similarity is greater than or equal to the audio threshold; If yes, set the value of participation condition 1 to "1"; If not, it is determined whether the difference between the audio similarity of other interactive terminals closest to the interactive terminal and the audio similarity is less than or equal to the difference threshold; when the difference is less than or equal to the difference threshold, the value of the first participation condition is set to "1"; otherwise, the value of the first participation condition is set to "0"; Step 6: Determine whether the interactive terminal participates in the activity based on participation condition 1 and participation condition 2.

2. A cross-platform multi-screen interaction method according to claim 1, characterized in that: The activity content includes interaction time, interaction actions and interaction rewards; the interaction time is the activity duration set by the organizer, the interaction action is the action to check whether the interactive terminal meets the second condition for participating in this activity, and the interaction reward is the reward given to the interactive terminal that completes this activity.

3. A cross-platform multi-screen interaction method according to claim 1, characterized in that: The audio similarity obtained by processing the activity simulation audio and the corresponding interactive collection audio includes: The interactive collected audio and the active simulated audio are sequentially preprocessed, feature extracted and audio signal aligned to obtain the comparative collected audio and the comparative simulated audio; the comparative collected audio and the comparative simulated audio are similarly calculated to obtain the audio similarity; wherein the preprocessing includes noise reduction and removal of repeated sounds; the feature extraction includes extraction of timbre features, pitch features and rhythm features; and the audio signal alignment uses dynamic time warping to perform audio alignment.

4. A cross-platform multi-screen interaction method according to claim 1, characterized in that: The step of judging whether the interactive terminal generates an interactive collection action and generating a second participation condition includes: Determining whether the interactive terminal generates an interactive collection action; If yes, set the value of participation condition 2 to "1"; If not, a confirmation message is sent to the corresponding interactive terminal, and it is determined whether the interactive terminal participates in the activity; when the interactive terminal participates in the activity, the corresponding interactive terminal is reminded to perform the interactive action, and the value of the second participation condition is set to "1"; otherwise, the value of the second participation condition is set to "0".

5. The cross-platform multi-screen interaction method according to claim 1, characterized in that: The determining whether the interactive terminal participates in the activity based on the first participation condition and the second participation condition includes: Determine whether the values ​​of participation condition 1 and participation condition 2 are both "1"; If yes, it is determined that the corresponding interactive terminal participates in the activity; If not, it is determined that the corresponding interactive terminal does not participate in the activity.