Bullet screen cloud game cooperative control method and device, equipment and medium

Abstract

The invention relates to the field of network live broadcast, and discloses a bullet screen cloud game cooperative control method and device, equipment and medium, and the method comprises the steps: receiving bullet screen instructions from a plurality of associated users of a cloud game, and synthesizing the bullet screen instructions into a live broadcast image of the cloud game in a text content form; when an organization instruction sent by a command user in the associated users is detected from the bullet screen instruction, determining a current game cooperation task according to the organization instruction; carrying out same-type combination on bullet screen instructions which are sent by each associated user in a life cycle corresponding to the game collaboration task and are matched with the game collaboration task, and constructing a group instruction; and acting the group instruction on a game process of the cloud game so as to present an action effect in the live broadcast image. According to the method, a new business mode is brought to the bullet screen cloud game, the processing efficiency of the game server and the user experience are remarkably improved, and the method has important application value and popularization significance.

Description

Technical Field

[0001] This application relates to network live broadcast technology, and in particular to a collaborative control method and device, equipment, and medium for barrage cloud games. Background Art

[0002] With the development of Internet technology, barrage cloud games, as a new form of games, have been loved by a large number of users. Barrage cloud games allow users in the game live broadcast room to participate in the game by sending barrage instructions. After these instructions are processed by the game server, the game process is changed and presented to the users through the live broadcast image. However, in the process of implementing barrage cloud games in traditional technologies, there are some technical problems that affect the user experience and system performance.

[0003] In traditional barrage cloud games, users participate in the game by sending barrage instructions, which usually appear in the game live broadcast image in text form. However, in this mode, users fight alone and lack effective organization and coordination. Each user sends barrage instructions based on their own thinking, resulting in the server needing to process each instruction of each user one by one. This processing method not only increases the burden on the server but also easily causes game delays and affects the real-time interaction experience of users.

[0004] Specifically, in traditional technologies, after receiving a barrage instruction, the server needs to parse the instruction content and convert it into specific operations in the game. This process involves a large amount of data processing and logical judgment. Especially when the number of users is large, the processing burden on the server will increase significantly. In addition, due to the lack of unified organization and coordination of the barrage instructions sent by users, the operations in the game often appear chaotic, affecting the fluency and fun of the game.

[0005] Furthermore, in traditional technologies, the barrage instructions sent by users usually have a high degree of freedom, and the instruction content may be relatively vague or unclear. When the server processes these instructions, it needs to perform additional semantic analysis and intention recognition, further increasing the complexity and delay of processing. In this case, the performance bottleneck of the server becomes one of the key factors restricting the development of barrage cloud games.

[0006] In summary, traditional barrage cloud game technologies have obvious technical problems in aspects such as user instruction processing, server burden, game delay, and user collaboration. These problems not only affect the user experience but also limit the further development of barrage cloud games. Therefore, it is necessary to propose a new technical solution to solve the above problems and improve the overall performance and user experience of barrage cloud games. Summary of the Invention

[0007] The purpose of this application is to solve the above problems and provide a collaborative control method for bullet screen cloud games, as well as its corresponding device, equipment, and non-volatile readable storage medium.

[0008] According to one aspect of the present application, a collaborative control method for bullet screen cloud games is provided, including the following steps:

[0009] Receive bullet screen instructions from multiple associated users of the cloud game, and synthesize the bullet screen instructions into the live image of the cloud game in the form of text content;

[0010] When an organization instruction sent by the commanding user among the associated users is detected from the bullet screen instructions, determine the current game collaboration task according to the organization instruction;

[0011] Merge the bullet screen instructions sent by each associated user within the life cycle corresponding to the game collaboration task and adapted to the game collaboration task into the same type, and construct them into a group instruction;

[0012] Apply the group instruction to the game process of the cloud game to present the action effect in the live image.

[0013] According to another aspect of the present application, a collaborative control device for bullet screen cloud games is provided, including:

[0014] An instruction receiving module, configured to receive bullet screen instructions from multiple associated users of the cloud game, and synthesize the bullet screen instructions into the live image of the cloud game in the form of text content;

[0015] A task determination module, configured to determine the current game collaboration task according to the organization instruction when an organization instruction sent by the commanding user among the associated users is detected from the bullet screen instructions;

[0016] An instruction integration module, configured to merge the bullet screen instructions sent by each associated user within the life cycle corresponding to the game collaboration task and adapted to the game collaboration task into the same type, and construct them into a group instruction;

[0017] An instruction application module, configured to apply the group instruction to the game process of the cloud game to present the action effect in the live image.

[0018] According to another aspect of the present application, a collaborative control device for bullet screen cloud games is provided, including a central processing unit and a memory. The central processing unit is used to call and run a computer program stored in the memory to execute the steps of the collaborative control method for bullet screen cloud games described in the present application.

[0019] According to another aspect of the present application, a non-volatile readable storage medium is provided, which stores a computer program implemented according to the above-mentioned collaborative control method for bullet screen cloud games in the form of computer-readable instructions. When the computer program is called and run by a computer, it executes the steps included in the method.

[0020] Through the innovative technology, the present application has achieved a technological innovation for bullet screen cloud games and obtained significant positive beneficial effects, including but not limited to:

[0021] First of all, in the process of processing bullet screen instructions, the present application introduces command users and organization instructions, making the cooperation between users more orderly and efficient. The command user can initiate organization instructions to clarify the current game collaboration tasks, and other users can perform collaborative operations according to these instructions to jointly complete the game collaboration tasks. This organized collaboration mode not only enhances the interaction and cooperation between users, but also brings more gameplay and fun to bullet screen cloud games, significantly expanding its business model.

[0022] Secondly, by merging the bullet screen instructions of multiple users into group instructions, the present application reduces the number of instructions that the game server needs to process. Under the drainage effect of the game collaboration task, the game server only needs to execute the group instructions corresponding to the game collaboration task, rather than processing each bullet screen instruction associated with the game collaboration task one by one, thus greatly reducing the burden on the game server, improving the processing efficiency of the game server, reducing the latency of the game, enabling users to participate in the game more smoothly, and enhancing the overall response efficiency.

[0023] In addition, based on the efficient user collaboration mode, the present application significantly improves the user experience of cloud games. Through organized collaborative operations, users can more clearly understand the game goals and tasks, thus participating in the game more effectively, enhancing the interaction and cooperation between users, making the game process more orderly and smooth, and improving the user's sense of participation and satisfaction. At the same time, due to the reduction of the burden on the game server and the improvement of the response efficiency, the problems of lag and latency encountered by users in the game are effectively solved, further improving the user experience. Description of the Drawings

[0024] Figure 1 A network architecture suitable for applying the collaborative control method for bullet screen cloud games of the present application is exemplary;

[0025] Figure 2 A flowchart of an embodiment of the collaborative control method for bullet screen cloud games of the present application;

[0026] Figure 3 A principle block diagram of the collaborative control device for bullet screen cloud games of the present application;

[0027] Figure 4 The figure is a schematic structural diagram of a collaborative control device for bullet screen cloud games adopted by this application. Specific embodiments

[0028] Before introducing the specific embodiments of the technical solution of this application in detail, the network architecture and application scenarios suitable for supporting the implementation of the technical solution of this application are first disclosed.

[0029] Please refer to Figure 1 , the exemplary network architecture of this application includes a game server 80, a live broadcast server 82, and terminal devices 90 of each user. The game server 80 is composed of multiple servers responsible for different specific services, and is responsible for providing cloud game services, including game logic processing, game state management, etc. The live broadcast server 82 is also realized by the division of labor of multiple specific servers, including a media server (RTC) responsible for live broadcast image processing and transmission and a service server responsible for front-end response to the terminal device 90, etc., for running a computer program product implemented according to the bullet screen cloud game collaborative control method of this application, and implementing each step of this method by collaborating with the game server 80.

[0030] The terminal devices 90 of each user can be various mobile terminals such as personal computers, smart phones, and tablets. These terminal devices 90 are mainly used to run a live broadcast application program, so that users can enter the live broadcast room through this application program, participate in cloud games provided by third parties under the control of the method of this application, and interact with other users. In this application, each live broadcast room usually includes a host user and audience users, and they can participate in the same cloud game together. Multiple live broadcast rooms can also participate in the same game together. All the audience users in each live broadcast room can form a group corresponding to that live broadcast room and participate in the same cloud game with other live broadcast rooms. Different opponent groups can also be formed within the same live broadcast room to participate in the same cloud game.

[0031] The screen of the cloud game can be pushed to the live broadcast room participating in the cloud game for playback. Specifically, the media server in the live broadcast server 82 processes the screen into a live stream and pushes it to the corresponding live broadcast room for playback, so that the live image viewed by the users in the live broadcast room contains the corresponding game screen.

[0032] In the live broadcast scene of the bullet screen cloud game, the bullet screen interaction between users is implemented as follows: after the user enters the live broadcast room through the terminal device 90, he can send bullet screen instructions to participate in the game. After these bullet screen instructions are received by the live broadcast server 82, they are processed by the business server and interact with the game server 80, thereby affecting the game process. For example, a user can send a bullet screen instruction to attack a character in the game, and these instructions will be processed by the game server 80, so that the game character will be attacked accordingly. At the same time, the media server in the live broadcast server 82 processes the game screen into a live stream and pushes it to the live broadcast room for playback, so that the user can watch the changes in the game screen in real time. In this way, the bullet screen interaction between users is realized, which enhances the interactivity and fun of the game.

[0033] To facilitate understanding of this application, the following is a preliminary description of various related concepts to be mentioned in this application:

[0034] "Associated users" refer to multiple users who participate in the same bullet screen cloud game, which can include both anchor users and audience users. These users enter the live broadcast room through the live broadcast application, participate in the cloud game together, and interact by sending bullet screen commands. Associated users can be friends or teammates in a temporary team. Together, they form a group in the game and can complete game tasks or compete together.

[0035] "Organization instructions" are bullet-screen instructions sent by users with specific permissions to clarify the current game collaboration tasks. Organization instructions usually contain specific guidance on the game process, such as "concentrate on defeating character A" or "gather resource X first". The purpose of organization instructions is to guide other related users to conduct organized collaborative operations to improve game efficiency and fun.

[0036] "Group instructions" are instructions that are combined by the server within the life cycle of the organization instructions by combining bullet screen instructions sent by multiple related users that are adapted to the collaborative tasks of the game. Group instructions contain the operation intentions and action degree data of multiple users, which can act on the game process more efficiently, reduce the processing burden of the server, and realize collaborative operations between users.

[0037] "Game collaborative tasks" refer to the current game goals or tasks determined by organizational instructions, such as "concentrate all efforts to defeat character A" or "complete resource collection together." Game collaborative tasks provide a common operating goal for related users, making the interaction between users more orderly and efficient.

[0038] See also Figure 2, A collaborative control method for bullet screen cloud games provided by this application can be implemented as a computer program product installed and running on a live server. In some embodiments of this method, the following steps are included:

[0039] Step S3100, Receive bullet screen instructions from multiple associated users of the cloud game, and synthesize the bullet screen instructions into the live image of the cloud game in the form of text content;

[0040] Receiving bullet screen instructions from multiple associated users of the cloud game and synthesizing these bullet screen instructions into the live image of the cloud game is the basis for realizing bullet screen cloud game interaction, ensuring that the instructions sent by users can be displayed in real time in the game live broadcast screen, and providing visual feedback for subsequent collaborative control.

[0041] Specifically, associated users enter the live broadcast room through their respective terminal devices and use the live broadcast application program to send bullet screen instructions. These instructions usually express the operation intentions of users in text form, such as attacking a certain game character, using a specific item, or completing a certain task. If an associated user sends a bullet screen instruction by voice, the corresponding audio data can also be converted into text form with the help of a speech recognition model. After receiving these bullet screen instructions, the live server can perform preliminary processing through the business server, including verifying the format and content of the instructions to ensure that they conform to the game logic and rules.

[0042] Furthermore, the media server is responsible for the mixing processing of the live image, and synthesizes these bullet screen instructions into the live image of the cloud game in the form of text content. During the mixing process, the media server assigns the text content of the bullet screen instructions in a predetermined style, with specific font, color, font size, etc., and superimposes it on the game screen. Moreover, a horizontal movement effect can also be given to these texts to play the role of automatically clearing the live image screen. For example, when a user sends a bullet screen instruction of "attack character A", this instruction will appear in a prominent text form such as bold and enlarged in a specified area of the game live broadcast screen, and other users can see this instruction in real time through the live image played in the graphical user interface of their terminal devices. To achieve this function, the live server can use real-time image processing technology to render the text content of the bullet screen instructions into an image layer and synthesize it with the game live broadcast stream.

[0043] In some embodiments, the display position of the bullet screen instructions can be adjusted according to the game scene and / or user preferences. For example, a dedicated bullet screen display area is set at the bottom or side of the game screen, or the display position is dynamically adjusted according to the urgency and importance of the instructions. In addition, the style of the bullet screen instructions (such as font size, color, transparency) can also be customized according to the game theme to enhance the visual effect and user experience.

[0044] The real-time reception and display of barrage instructions provide an intuitive interaction platform for users, enabling them to participate in the game process by sending barrage instructions and observe the operation intentions of themselves and other users in real time.

[0045] Step S3200: When an organization instruction sent by the commanding user among the associated users is detected from the barrage instruction, determine the current game collaboration task according to the organization instruction;

[0046] After receiving the barrage instruction, the live broadcast server can first detect it to determine whether it belongs to the specific organization instruction of this application. When an organization instruction sent by the commanding user among the associated users is detected from the barrage instruction, the current game collaboration task can be further determined according to the organization instruction, so that through the effect of the game collaboration task, orderly collaborative operations can be carried out according to the organization intention of the user.

[0047] Specifically, the organization instruction is a barrage instruction sent by a commanding user with specific permissions, and its purpose is to clarify the current game collaboration task. The commanding user can be the host user or an audience user who obtains permissions through a specific method (such as virtual gift giving). These users can send organization instructions to guide other associated users to carry out collaborative operations, thereby realizing more orderly game interactions. The organization instruction usually contains operation intentions and targets. When the live broadcast server receives a barrage instruction, it will first detect whether the instruction comes from a user with commanding permissions and determine whether it is an organization instruction. To efficiently determine the game collaboration task from the organization instruction, this application provides but is not limited to the following two embodiments.

[0048] The first embodiment is to directly extract the keywords defining the game collaboration task from the text content of the organization instruction based on preset rules. For example, in the organization instruction "Concentrate forces to attack character A", "Attack character A" is the keyword, which directly defines the current game collaboration task as "Attack character A". The preset rules can be implemented through keyword lists, semantic analysis models, regular expressions, etc., and are used to quickly and accurately extract key information from the text content. In this way, the core intention of the organization instruction can be quickly identified and transformed into a specific game collaboration task.

[0049] The second embodiment is to input the live image synchronized with the timestamp and the text content of the organization instruction into a neural network model with multimodal reasoning ability, such as a large language model, to determine the corresponding game collaboration task. The multimodal reasoning model can process text and image data simultaneously. By analyzing the game scene in the live image and the text content of the organization instruction, it can more accurately understand the intention of the organization instruction. For example, when the organization instruction is "attack character A", the multimodal reasoning model can combine information such as the position and status of character A in the live image to further confirm the game operation target, that is, the attack target and the game operation type. This embodiment can handle more complex and ambiguous organization instructions, improving the accuracy and flexibility of task recognition.

[0050] The second embodiment has greater advantages in terms of intelligence. For example, when the organization instruction is "attack the enemy in front", the second embodiment can be used to identify the enemy in front in the live image and use it as the game operation target. In addition, relying on the technical advantages of the second embodiment, it is also possible to dynamically adjust the game collaboration task according to the game rules and scenarios. For example, in a team battle scenario, if the organization instruction is "protect the teammates", according to the position and status of the teammates in the live image, the task content will be understood as "attack the enemy approaching the teammates" or "provide support for the teammates", and the game collaboration task will be determined accordingly.

[0051] Regardless of which embodiment is adopted, the game collaboration task mainly includes the game operation type and the game operation target. The game operation type refers to the specific operations that the user needs to perform in the game, such as attack, defense, collection, etc.; the game operation target is the specified target object or target task, such as a specific game character, item, area, indicator, etc. By clarifying the game operation type and target, the organization instruction can be transformed into a specific game collaboration task, guiding other associated users to perform organized game operations.

[0052] For example, in a multiplayer online bullet screen cloud game, the commander user sends an organization instruction "concentrate on attacking character A". Through preset rules or a neural network model, the game collaboration task is identified as "attack character A", where the game operation type is "attack" and the game operation target is "character A". Subsequently, this game collaboration task is conveyed to other associated users, and bullet screen instructions suitable for this task are collected within the preset life cycle for subsequent collaborative operations.

[0053] In a specific embodiment, multiple ways can be adopted to determine the game collaboration task. For example, if the text content of the organization instruction contains clear game operation objectives and game operation types, the server can directly use this information as the task content according to the first embodiment. If the task content cannot be extracted according to the first embodiment, that is, the organization instruction is relatively ambiguous, then according to the second embodiment, a comprehensive judgment can be made by combining the text content and the live image.

[0054] Step S3300: Merge the barrage instructions sent by each associated user within the corresponding life cycle of the game collaboration task and adapted to the game collaboration task into the same type, and construct them into a group instruction.

[0055] The life cycle of the organization instruction refers to the time period from when the commanding user sends the organization instruction to generate a game collaboration task for application until the game collaboration task corresponding to the instruction is completed or terminated. The life cycle of the organization instruction can be implemented in multiple ways, specifically depending on factors such as the complexity of the game scenario, the urgency of the task, and the number of users. For example, in a fast-paced battle scenario, the life cycle can be set to a shorter time, such as 5 seconds, to ensure a quick response; while in a scenario that requires strategic planning, the life cycle can be extended to several minutes. In addition, the life cycle can also be dynamically adjusted according to specific events in the game process. For example, when the target in the game is defeated or the task is completed, the life cycle ends. In some embodiments, the duration of the life cycle can also be determined according to the permission level of the commanding user or the cumulative consideration of the virtual gifts given by the user to the host user. For example, a user with a higher permission level or a user who gives more virtual gifts can have a longer life cycle to reflect their greater influence on the game process. In this way, the life cycle of the organization instruction can flexibly adapt to different game scenarios and user needs, thereby optimizing the collaborative control effect of the game.

[0056] The live server can perform time slicing on the life cycle through a preset time slot, and regularly merge the barrage instructions sent by each associated user within the corresponding life cycle of the game collaboration task and adapted to the game collaboration task into the same type, and construct a group instruction. The time slot corresponding to regularly constructing the instruction is usually much smaller than the life cycle. For example, when the life cycle is 5 seconds, this step can be regularly executed according to 0.5 seconds to quickly construct a group instruction for application, avoiding the user's perception of delay. It can be seen that the group instruction is comprehensively applied by integrating the instructions of multiple users, reducing the server processing burden, and at the same time improving the game response efficiency and user experience.

[0057] After receiving the barrage instructions sent by the associated users during the life cycle, the live broadcast server first parses and identifies these instructions to determine whether they are suitable for the current game collaboration task. The judgment basis for suitability includes the text content of the instructions, the type of game operation indicated by the instructions, and / or the game operation target of the instructions, etc. For example, if the game collaboration task is "attack character A", then all barrage instructions pointing to attacking character A will be regarded as suitable instructions. Another example is that when the game operation target is missing in the barrage instruction, such as the specified character is missing, but the content "attack him" is given, it can also be directly regarded as a barrage instruction dominated by the game collaboration task and considered the same instruction as "attack character A".

[0058] In some embodiments, if the game operation target specified by the barrage instruction is different from the game operation target of the game collaboration task, such a barrage instruction is not suitable for the game collaboration task, and for the unsuitable instructions, they will not be incorporated into the construction of the group instructions. In this way, during the period when the game collaboration task is in effect, it not only allows the associated users to carry out autonomous actions by sending barrage instructions with different game operation targets, but also allows most of the clear or ambiguous barrage instructions of the associated users to be parsed into the same instructions that comply with the game collaboration task, which not only improves the instruction execution efficiency but also does not affect the decoupling and cooperation between the associated users.

[0059] After determining the suitable barrage instructions, the live broadcast server can further determine the game operation type and action degree data of the corresponding associated users according to the text content of these instructions. The game operation type refers to the specific operations that the user needs to perform in the game, such as attack, defense, collection, etc.; the action degree data represents the intensity or degree of the operation, for example, the strength of the attack, the duration, or the number of items used, etc. Through these data, the server can more accurately understand the user's operation intention and convert it into executable game instructions. Specifically, when implementing, for the unclear content in the barrage instruction, a preset default value can be assigned. For example, if the barrage instruction only gives "knock it down", here only the game operation type is given in the context, but the specific action degree data such as the attack strength data is not given. Therefore, the corresponding default value can be used as the action degree data as the parameter specified by the corresponding associated user.

[0060] To improve the instruction execution efficiency and reduce the system overhead of the game server, the live server will classify the barrage instructions of each associated user according to the game operation type, obtaining each category corresponding to different game operation types. For each category of barrage instructions, the server will accumulate the action degree data thereof to obtain the overall degree data. For example, if multiple users send barrage instructions to attack character A, the server will accumulate the action degree data (such as attack strength) of these instructions to obtain an overall attack strength value. Then, the server will construct corresponding group instructions according to the game operation type, the overall degree data of each category of barrage instructions, and the game action target specified in the game collaboration task.

[0061] There can be multiple specific implementation manners for the construction process of the group instructions. For example, in a simple embodiment, if multiple users send barrage instructions to attack character A, the server can merge these instructions into a group instruction, whose game operation type is "attack", the overall degree data is the accumulated attack strength value, and the game action target is "character A". In a more complex embodiment, the server can consider factors such as the time sequence of the instructions, the privilege level of the users, the urgency of the instructions, etc., to construct the group instructions more precisely. For example, for the instructions sent by high-privilege users, higher weights can be given; for urgent instructions, they can be processed preferentially.

[0062] In this case, during the life cycle of the same game collaboration task, multiple group instructions corresponding to different game operation types can be generated concurrently, and these group instructions may be applied successively or synchronously. It should be noted that when the game task target is achieved, the life cycle can be directly ended. For example, if character A in the above example is knocked down, the life cycle of the corresponding game collaboration task will naturally terminate, and the game collaboration task can be destroyed, and the subsequent barrage instructions initiated by the associated users will no longer be associated with this game collaboration task.

[0063] Through this method of merging the same type, the live server can efficiently integrate the barrage instructions of multiple users into group instructions, reducing the number of instructions for the game server, improving the response speed and fluency of the game. At the same time, the construction of the group instructions also makes the operations in the game more orderly and efficient, enhancing the collaborative effect among users and improving the overall game experience.

[0064] Step S3400: Apply the group instructions to the game process of the cloud game to present the action effect in the live image.

[0065] The live server applies the constructed group instructions to the game process of the cloud game and presents the effect of the application in the live image, so as to achieve the collaborative control and real-time feedback of the bullet screen cloud game, ensuring that the bullet screen instructions sent by users can efficiently affect the game process in the form of group instructions and be real-time fed back to all associated users.

[0066] Specifically, after constructing the group instructions, the live server sends the group instructions to the game server of the cloud game through a preset interface. After receiving the group instructions, the game server performs corresponding operations on the game process according to the game operation type, overall degree data, and game application target in the instructions. For example, if the group instruction is "focus on attacking character A", the game server will attack character A according to the accumulated attack strength value and update the game state in real time.

[0067] In some embodiments, to ensure that the effect of the application can be presented to all associated users in real time, after executing the group instructions, the game server returns the corresponding action effect data. This data includes the result of the operation (such as the current health value of character A, whether it is defeated, etc.) and the visual effects generated by the operation (such as attack animations, special effects, etc.). After receiving this action effect data, the live server constructs corresponding action special effects according to the data and synthesizes these special effects into the live image through the media server. For example, when character A is attacked, the corresponding attack animation and special effects will appear in the live image, and at the same time, the change in the health value of character A will be displayed.

[0068] In some embodiments, the media server can adopt a variety of technical means to optimize the presentation of action effects. For example, real-time image processing technology can be used to render the action special effects as an image layer and synthesize it with the game live stream. In addition, the display style of the special effects can also be adjusted according to the game scene and user preferences. For example, a prominent light effect is displayed when the attack is successful, and an explosion special effect is displayed when the character is defeated.

[0069] In this way, the live server can not only efficiently apply the group instructions to the game process, but also ensure that all associated users can view the operation effects in real time, enhancing the interactivity and interest of the game. For example, in a multiplayer online bullet screen cloud game, after the commander user sends the organizational instruction of "focus on attacking character A", the attack instructions sent by multiple associated users are merged into a group instruction. After the game server executes this group instruction, character A is defeated, and the special effects and animations of character A being defeated are displayed in the live image. All associated users can see this effect in real time, thus enhancing the sense of cooperation and participation in the game.

[0070] In addition, the live server can also dynamically adjust the presentation method of action effects according to specific events in the game process. For example, when the target in the game is defeated or the task is completed, special celebration effects can be displayed; when the game enters a new stage, the style of the effects can be adjusted to match the new game scene. In this way, the live server can flexibly adapt to different game scenes, further enhancing the visual effects and user experience of the game.

[0071] Through the above embodiments, the present application realizes the technological innovation of bullet screen cloud games and achieves significant positive beneficial effects, including but not limited to:

[0072] First of all, through innovative business logic, a new user collaboration mode is brought to cloud games, significantly expanding the business model of bullet screen cloud games. In traditional bullet screen cloud games, although users can participate in the game by sending bullet screen instructions, these instructions often lack effective organization and coordination, resulting in chaotic operations in the game, affecting the fluency and fun of the game. However, the present application makes the collaboration between users more orderly and efficient by introducing command users and organizing instructions. The command user can initiate an organization instruction to clarify the current game collaboration task, and other users can perform collaborative operations according to these instructions to jointly complete the game collaboration task. This organized collaboration mode not only enhances the interaction and cooperation between users, but also brings more gameplay and fun to bullet screen cloud games, thus expanding its business model.

[0073] Secondly, the present application significantly reduces the burden on the game server, helping to improve the response efficiency of bullet screen cloud games. In traditional technologies, the game server needs to process each bullet screen instruction of each user one by one. In the case of a large number of users, this will cause a heavy processing burden on the game server, thereby affecting the response speed and user experience of the game. However, the present application reduces the number of instructions that the game server needs to process by merging the bullet screen instructions of multiple users into group instructions. Under the drainage effect of the game collaboration task, the game server only needs to execute the group instructions corresponding to the game collaboration task, without processing each bullet screen instruction associated with the game collaboration task one by one, thus greatly reducing the burden on the game server. This optimization not only improves the processing efficiency of the game server, but also reduces the latency of the game, enabling users to participate in the game more smoothly and improving the overall response efficiency.

[0074] In addition, based on an efficient user collaboration model, this application significantly enhances the user experience of cloud games. Through organized collaborative operations, users can more clearly understand game goals and tasks, and thus participate in the game more effectively. This collaboration model not only enhances the interaction and cooperation among users, but also makes the game process more orderly and smooth, improving the users' sense of participation and satisfaction. At the same time, due to the reduction of the game server load and the improvement of the response efficiency, the problems of lag and delay encountered by users in the game are effectively solved, further enhancing the user experience.

[0075] Based on any embodiment of the method of this application, determining the current game collaboration task according to the organization instruction includes:

[0076] Step S3210: Analyze the organization instruction to determine its corresponding text content;

[0077] After receiving the organization instruction, the live server starts to analyze the organization instruction to determine its corresponding text content. The organization instruction is usually sent in text form and contains specific guidance for directing users' game progress. For example, the organization instruction may be "Concentrate on attacking character A" or "Prioritize collecting resource X". By analyzing this organization instruction, the corresponding text content can be directly extracted. However, this text content may not be clear enough in meaning and thus cannot be directly used as the task content of the game collaboration task. The text content extracted in this step is basically the original text when the user inputs, and some stop words, symbols, emoticons, etc. input by the user can be directly deleted.

[0078] Step S3220: Obtain multiple consecutive image frames including the image frame corresponding to the timestamp of the organization instruction in the live image of the cloud game according to the timestamp of the organization instruction;

[0079] The timestamp of the organization instruction is used to synchronize the instruction with the game live screen to ensure the real-time and accuracy of task recognition. The live server obtains the image frame corresponding to the timestamp of the organization instruction and multiple consecutive image frames before and after it from the live stream of the cloud game according to the timestamp of the organization instruction. For example, if the timestamp of the organization instruction corresponds to a certain frame in the live image, the server will obtain this frame and 5 frames before and after it, forming an image sequence containing 11 frames. These consecutive image frames can provide richer game scene information to help more accurately understand the intention of the organization instruction. The process of obtaining the image frames can be realized through real-time image processing technology to ensure that the acquisition and processing of the image frames are completed within milliseconds to meet the real-time requirements of the game. Step S3230: Perform task recognition according to the text content and the multiple consecutive image frames to determine the game collaboration task corresponding to the organization instruction.

[0080] The live server first performs semantic analysis on the text content of the organization instruction to extract key information such as the operation intention and the target object. For example, for the organization instruction "Concentrate forces to attack Character A", the semantic analysis model can identify "attack" as the operation intention and "Character A" as the target object. This process can be achieved through techniques such as a preset keyword list, a semantic analysis model, or regular expressions to ensure the rapid and accurate extraction of key information from the text.

[0081] Meanwhile, the live server uses multiple consecutive image frames obtained for image analysis to acquire real-time information in the game scene. For example, through image recognition technology, the server can determine the position and status of Character A in the game (such as whether it is in a combat state) and other relevant environmental information. The combination of this image information and the text content can more accurately understand the intention of the organization instruction and transform it into specific game collaboration tasks.

[0082] In some embodiments, the live server can adopt a multimodal reasoning model, such as a large language model combined with image recognition technology, to comprehensively analyze the text content of the organization instruction and the image frames. For example, when the organization instruction is "Attack the enemy in front", the multimodal reasoning model can combine the position and status information of the enemy in front in the image frame to accurately identify the attack target and determine the game collaboration task as "Attack the enemy in front" accordingly.

[0083] In addition, the live server can also dynamically adjust the logic of task recognition according to the game rules and the scene. For example, in a team battle scene, if the organization instruction is "Protect teammates", the server can analyze the position and status of the teammates through image frames, understand the task content as "Attack the enemies approaching the teammates" or "Provide support to the teammates", and determine the game collaboration task accordingly.

[0084] Through this task recognition method that combines text content and image frames, the live server can accurately transform the organization instruction into specific game collaboration tasks, guiding other associated users to perform organized game operations. For example, in a multiplayer online bullet screen cloud game, the commanding user sends the organization instruction "Concentrate forces to attack Character A". The server determines the game collaboration task as "Attack Character A" through semantic analysis and image recognition technology, conveys this task to other associated users, and simultaneously collects bullet screen instructions adapted to this task within a preset lifecycle for subsequent collaboration operations.

[0085] This embodiment significantly improves the accuracy and efficiency of organization instruction parsing by comprehensively using a method combining multimodal data. After receiving the organization instruction, the live broadcast server not only performs semantic analysis on the instruction text to extract key information, but also makes a comprehensive judgment in combination with the real-time game scene information in the image frame. This multimodal reasoning method can effectively solve the problem of fuzzy or unclear meaning of text content and ensure the accuracy of instruction parsing. At the same time, using the game scene information in the image frame, the server can more accurately identify the operation target and environmental state, thereby accurately converting the organization instruction into a specific game collaborative task. In addition, by dynamically adjusting the task recognition logic to adapt to different game scenes and rules, this embodiment further enhances the flexibility and adaptability of the system. This comprehensive multimodal data processing method not only improves the speed and accuracy of instruction parsing, but also reduces the erroneous operation caused by unclear instructions, significantly improves the real-time and interactivity of the game, and provides users with a smoother and more orderly game experience.

[0086] On the basis of any embodiment of the method of the present application, the bullet screen instructions sent by each associated user within the corresponding life cycle of the game collaborative task and adapted to the game collaborative task are merged into a group instruction of the same type, including:

[0087] Step S2100, parsing each bullet screen instruction of the plurality of associated users generated in the life cycle, and extracting the text content thereof;

[0088] The server parses each bullet message command and extracts its text content. It mainly performs semantic analysis on the original text input by the user, removes meaningless stop words, symbols or emoticons, and obtains clear command intent. For example, the user may send commands such as "attack character A" or "support teammates". The server extracts key text content through parsing, laying the foundation for subsequent task direction recognition.

[0089] Step S2200: Perform task orientation identification on each of the text contents to determine the type of collaboration intention, so as to determine whether the corresponding bullet screen instruction is suitable for the game collaboration task;

[0090] The server performs task - pointing recognition on the text content of each barrage instruction to determine its type of collaboration willingness. In this embodiment, the type of collaboration willingness is divided into an independent execution type and a non - independent execution type. Instructions of the independent execution type refer to operations that the user hopes to complete independently, without relying on the collaboration of other users and not participating in the game collaboration task; while instructions of the non - independent execution type indicate that the user is willing to participate in the game collaboration task, and their operations need to be merged with the instructions of other associated users. For example, the user - sent "Attack Character B" is significantly different from the game collaboration task of "Attack Character A" and belongs to the independent execution type, while "Attack A" belongs to the non - independent execution type dominated by this game collaboration task.

[0091] Step S2300: When the type of collaboration willingness belongs to the independent execution type, directly apply the corresponding barrage instruction to the game process of the cloud game alone;

[0092] For barrage instructions of the independent execution type, the server transmits them to the game server by calling a preset interface by providing corresponding parameters. The game server executes the corresponding operation instructions and directly applies them to the game process of the cloud game without merging processing. This method ensures that the user's independent operations can be executed quickly, meeting the user's need for instant feedback. For example, when a user discovers an independent target in the game scene and hopes to attack it quickly, the independent execution instruction sent by the user will be processed immediately without affecting the collaborative operations of other users.

[0093] Step S2400: When the type of collaboration willingness belongs to the non - independent execution type, determine that the corresponding barrage instruction is adapted to the game collaboration task for constructing a group instruction.

[0094] For barrage instructions of the non - independent execution type, the server marks them as instructions adapted to the game collaboration task for subsequent construction of group instructions. These instructions will be collected and merged to form a group instruction, thereby realizing the collaborative operations of multiple users. For example, after the organizer commands the users to issue an organizational instruction of "Concentrate on attacking Character A", when multiple users send instructions such as "Attack Character A" and "Use Prop M on him", they will be recognized as non - independent execution types and merged into a group instruction to enhance the attack strength on Character A.

[0095] This embodiment significantly improves the collaborative control efficiency and user experience in the bullet screen cloud game by accurately parsing and classifying the bullet screen instructions of the associated users. First, by performing semantic analysis and task orientation identification on the text content of the bullet screen instructions, the server can accurately distinguish between autonomous execution types and non-autonomous execution types of instructions. This distinction mechanism ensures that the user's independent operation instructions can quickly and directly act on the game process to meet the needs of instant feedback, and at the same time efficiently merges the instructions adapted to the collaborative tasks of the game into group instructions to achieve collaborative operations of multiple users. This processing method not only optimizes the server's instruction processing flow and reduces unnecessary processing burdens, but also enhances the real-time and interactivity of the game, making the game process smoother and the execution of collaborative tasks more efficient. In addition, through this refined instruction management, this embodiment also improves the overall collaborative effect of the game, making the interaction between users more orderly and efficient, and further enhancing the fun and sense of participation in the bullet screen cloud game.

[0096] On the basis of any embodiment of the method of the present application, the bullet screen instructions adapted to the game collaborative task sent by each associated user within the corresponding life cycle of the game collaborative task are merged into a group instruction of the same type, including:

[0097] Step S3310: determining the game operation type and action degree data of the corresponding associated user according to the text content of the bullet screen instruction adapted to the game collaborative task, wherein the game collaborative task includes a game operation target and a game operation type;

[0098] Game collaborative tasks usually include game operation targets and game operation types. Game operation targets refer to the specific objects operated by users, such as a game character, item or area, or the task goals that users hope to achieve, such as defeating a character or achieving a certain degree of killing effect; game operation types refer to the specific operations that users need to perform in the game, such as attack, defense, collection, movement, etc. For example, in a game scene, if the game collaborative task is "concentrate on attacking character A", then "character A" is the game operation target, and "attack" is the game operation type.

[0099] For each bullet screen instruction adapted to the game collaborative task, the server extracts its game operation type and action degree data through text parsing technology. The action degree data can be the intensity, frequency, duration or other relevant parameters of the user's operation. For example, the bullet screen instruction sent by the user may be "attack character A with a heavy blow", and the server can parse out that the game operation type is "attack" and the action degree data is "heavy blow". If the instruction is "continue to attack character A for 10 seconds", the action degree data may include "duration 10 seconds".

[0100] In some embodiments, the server can use a preset semantic analysis model to parse the text content of the barrage instructions. These models can be implemented based on keyword matching, natural language processing techniques, or deep learning algorithms. For example, through keyword matching, the server can identify operation types such as "attack", "support", "move", etc., and action degree descriptions such as "heavy strike", "quick", "continuous", etc. For more complex instructions, such as "attack character A with a flame skill", the server can combine natural language processing techniques to extract the game operation type as "attack" and the action degree data as "flame skill".

[0101] In addition, the server can also assign default action degree data to unclear instructions according to the text content of the barrage instructions. For example, if the user only sends "attack character A" without specifying the attack method or intensity, the server can adopt a preset default value, such as "normal attack" or "standard intensity".

[0102] Step S3320: Classify the barrage instructions of each associated user according to the game operation type, and accumulate the action degree data of each barrage instruction in each category to obtain the overall degree data;

[0103] After the server determines the game operation type of each barrage instruction, it classifies the instructions with the same operation type into one category. For example, all "attack" type instructions are classified into one category, and all "support" type instructions are classified into another category. For each category of instructions, the server further processes its action degree data. The action degree data can be specific parameters such as the intensity, duration, and frequency of the attack. For example, for "attack" type instructions, the server will accumulate the attack intensity and duration of each instruction to obtain an overall attack intensity and total duration. If a user sends "attack character A with a heavy strike" and another user sends "continuously attack character A for 10 seconds", the server will accumulate these action degree data to obtain the overall degree data for forming a stronger attack instruction.

[0104] In some embodiments, the server can use multiple methods to process the action degree data. For example, for the attack intensity, the server can quantify it as a numerical value, such as "heavy strike" corresponding to a higher value and "light strike" corresponding to a lower value. For the duration, the server can directly accumulate the time values of each instruction. In addition, the server can also perform weighted processing on the action degree data according to the user's permission level or the urgency of the instruction. For example, instructions sent by high-privilege users may be given a higher weight, so that they occupy a larger proportion when accumulating.

[0105] Step S3330: Construct corresponding group instructions for each type of barrage instruction. The group instructions include the game operation type corresponding to this type of barrage instruction, the overall degree data, and the game operation target specified in the game collaboration task.

[0106] Group instructions are instructions obtained by integrating the action degree data of multiple users under the same operation type, and are used to act on the game process efficiently. Group instructions ensure that the collaborative operations between associated users can be achieved in a more efficient and orderly manner.

[0107] The group instructions in this embodiment include but are not limited to the following elements: game operation type, overall degree data, and game operation target. The game operation type refers to the specific operations that users need to perform in the game, such as attacking, defending, collecting, etc.; the overall degree data is obtained by accumulating the action degree data of each adapted barrage instruction, and reflects the intensity, frequency, or duration of the user's operations, etc.; the game operation target is the specific object or task target specified in the game collaboration task, such as attacking a certain character or completing a certain task.

[0108] Accordingly, when constructing the corresponding group instructions for each game operation type, in accordance with the preset instruction format, encapsulate the game operation type, its corresponding overall degree data, and the game operation target into the same instruction.

[0109] It should be noted that during the life cycle of the game collaboration task, each step of this embodiment can be continuously and cyclically executed at intervals of preset time slots, so as to not only respond to the barrage instructions of each associated user in a timely manner, but also minimize the number of instructions processed by the server.

[0110] In this embodiment, by merging barrage instructions adapted to game collaboration tasks of the same type and constructing group instructions, the collaborative control efficiency and user experience of barrage cloud games are significantly improved. First, by parsing the text content of the barrage instructions, the game operation type and action degree data are accurately extracted, ensuring the accuracy and executability of the instructions. Secondly, the instructions are classified according to the game operation type, and the action degree data is accumulated to generate the overall degree data. This process effectively integrates the operation intentions of multiple users, reduces the number of instructions processed by the server, reduces the burden on the server, and improves the response speed and game fluency. Finally, the constructed group instructions include the game operation type, the overall degree data, and the game operation target, and can act on the game process efficiently to achieve collaborative operations among users. In addition, by presetting time slots to execute these steps cyclically, the instructions of users can be responded to in a timely manner, and at the same time, the number of instructions processed by the server is minimized to further optimize the instruction processing flow and game performance. This refined instruction management and collaborative operation mechanism not only enhances the real-time and interactivity of the game, but also improves the user's sense of participation and satisfaction, bringing a more efficient and orderly collaborative experience to barrage cloud games.

[0111] Based on any embodiment of the method of the present application, before merging barrage instructions sent by each associated user within the corresponding life cycle of the game collaboration task and adapted to the game collaboration task of the same type and constructing them into group instructions, it includes:

[0112] Step S4100, applying a preset policy to determine the duration corresponding to the life cycle of the latest organization instruction according to the cumulative consideration of the virtual gifts given by the commanding user to the host user;

[0113] By applying a preset policy, the live server can dynamically adjust the effective duration of the organization instruction through the cumulative consideration of the virtual gifts given by the commanding user to the host user, thereby optimizing the collaborative control effect of the game.

[0114] Specifically, the cumulative consideration of virtual gifts refers to the total value of all virtual gifts given by the commanding user to the host user within a certain period of time. These virtual gifts can be in-game currency, items, or other forms of rewards. The server associates the cumulative consideration of virtual gifts with the life cycle duration of the organization instruction through a preset policy. For example, the preset policy can stipulate that when the cumulative consideration reaches a certain threshold, the life cycle duration of the organization instruction increases accordingly; conversely, if the cumulative consideration is low, the life cycle duration may be short.

[0115] There can be various specific implementation manners for the preset strategy. For example, in one embodiment, the server can set a linear relationship formula, that is, the higher the cumulative consideration of the virtual gift, the longer the life cycle duration of the organization instruction. In another embodiment, the server can adopt a piecewise function, dividing the cumulative consideration into different intervals, and each interval corresponds to a different life cycle duration. For example, when the cumulative consideration is between 100 and 200 game coins, the life cycle is 5 seconds; when the cumulative consideration is between 200 and 300 game coins, the life cycle is 10 seconds, and so on.

[0116] In addition, the server can also dynamically adjust the preset strategy according to the specific scenarios and rules of the game. For example, in a fast-paced battle scenario, even if the cumulative consideration of the virtual gift is high, the life cycle of the organization instruction may be set to a shorter time to ensure the fast response of the game; while in a scenario with stronger strategy, the life cycle may be extended accordingly to allow users to perform more complex collaborative operations.

[0117] In this way, the live server can dynamically adjust the life cycle of the organization instruction according to the contribution of the commanding user (reflected by the cumulative consideration of the virtual gift), so as to better balance the interactivity and fluency of the game. This method not only improves the flexibility and adaptability of the game, but also provides users with a richer and fairer game experience.

[0118] Step S4200: Detect whether the permission level of the associated user who issues the latest organization instruction exceeds the permission level of the associated user to whom the currently applied organization instruction belongs. When it exceeds, end the currently applied organization instruction and send a notification message to each associated user, and enable the latest organization instruction to make its life cycle take effect.

[0119] Generally, if there is no currently applied organization instruction, then the newly generated organization instruction by the system should be immediately applied. But sometimes there will inevitably be a situation where multiple associated users compete for the command right of the organization instruction. In response to this situation, the live server detects whether the permission level of the associated user who issues the latest organization instruction exceeds the permission level of the associated user to whom the currently applied organization instruction belongs, and conducts conflict management accordingly to ensure the efficient and orderly execution of the game collaboration task, and optimizes the collaborative control effect of the game by dynamically adjusting the priority of the organization instruction.

[0120] Specifically, each associated user has a corresponding permission level, which can be determined in advance based on the user's role in the game, contributions, virtual gift giving, or other factors. For example, a commanding user may have a higher permission level due to their organizational role in the game, while the permission level of an ordinary viewer user is relatively low. When an associated user sends a new organizational instruction, the server compares the permission level of this user with the permission level of the user to whom the currently applied organizational instruction belongs. If the sender of the new instruction has a higher permission level, it indicates that they have a greater influence on the game process, or their instruction may be more critical to the synergy effect of the game.

[0121] In this case, the server ends the currently applied organizational instruction and sends a notification message to all associated users, informing them that the new organizational instruction is about to take effect. This notification message can include information such as the specific content of the new instruction and the effective time, ensuring that all users can timely understand the changes in the game synergy tasks. Subsequently, the server enables the latest organizational instruction, making its life cycle start to take effect, thereby guiding the game process to proceed according to the new synergy tasks.

[0122] This dynamic adjustment mechanism can not only flexibly adjust the game synergy tasks according to the user's permission level, but also ensure the coherence and fairness of the game process, and coordinate the competitive relationship between various organizational instructions. For example, in a team battle scenario, if a high-permission commanding user sends a new organizational instruction, such as "Concentrate fire on character B", while the currently executed instruction is "Protect character A", the server will decide whether to switch to the new instruction based on the comparison result of the permission levels. If the switch occurs, all associated users will receive a notification and start performing operations according to the new synergy tasks.

[0123] In this embodiment, by dynamically adjusting the lifecycle and priority of organization instructions, the collaborative control efficiency and user experience of bullet screen cloud games are significantly improved. First, by associating the cumulative consideration of virtual gifts commanded by users to the host user with the lifecycle duration of organization instructions, the server can dynamically adjust the effective duration of instructions according to the contributions of users. This mechanism not only encourages users to actively participate in game interactions, but also optimizes the collaborative effect of the game by flexibly adjusting the instruction duration, ensuring that the game process can respond quickly according to real-time situations. Second, by detecting the permission level of new organization instructions and deciding whether to switch instructions accordingly, the server can ensure the efficient execution of game collaborative tasks while maintaining the coherence and fairness of the game process. This dynamic adjustment mechanism enables the game to flexibly switch collaborative tasks according to the influence of users and the importance of instructions, avoiding the decline in game experience caused by instruction conflicts. In addition, by sending notification messages to all associated users, it ensures that users are informed in a timely manner of changes in game collaborative tasks, enhancing the sense of participation and satisfaction of users. Generally speaking, in this embodiment, through refined instruction management and dynamic adjustment mechanisms, not only the flexibility and adaptability of the game are improved, but also a richer and fairer game experience is provided for users, significantly enhancing the interactivity and smoothness of bullet screen cloud games.

[0124] Based on any embodiment of the method of the present application, receiving bullet screen instructions from multiple associated users of a cloud game and synthesizing the bullet screen instructions into the live image of the cloud game in the form of text content includes:

[0125] Step S3110, receiving a virtual gift giving request submitted by any associated user, and transferring the virtual gift carried by the request from the personal account of the associated user to the personal account of the host user;

[0126] The live server can receive a virtual gift giving request corresponding to a gift given by any associated user to the host user in the live broadcast room, and correspondingly execute the transfer operation of the virtual gift. Through the giving of virtual gifts, not only the interactivity between users is enhanced, but also a basis is provided for subsequent permission level setting and execution of collaborative tasks.

[0127] Specifically, after an associated user enters the live broadcast room through a terminal device, a virtual gift giving request can be submitted through a live broadcast application program. These virtual gifts can be in-game currency, items or other forms of rewards, and their value is usually expressed in in-game currency units or other quantitative forms. When an associated user submits a virtual gift giving request, the live server receives the request and transfers the corresponding virtual gift from the personal account of the associated user to the personal account of the host user. This transfer process can be achieved by calling a preset payment interface or account management module to ensure the safe and accurate transfer of virtual gifts.

[0128] In some embodiments, virtual gifts can be presented in various ways. For example, a user can click on the virtual gift icon in the live streaming interface, select the type and quantity of the gift to be presented, and then confirm the presentation operation. After receiving the request, the server will verify whether the user's account balance is sufficient. If it is sufficient, the transfer operation will be completed; if it is insufficient, an error message will be returned to the user. In addition, the server can also record the detailed information of each virtual gift presentation, including the presentation time, gift type, quantity, sender, and recipient, etc., for subsequent statistical analysis and dynamic adjustment of permission levels.

[0129] Step S3120: Set the permission level of the associated user corresponding to the type of virtual gift presented by the associated user to the host user.

[0130] The live streaming server sets the permission level of the associated user corresponding to the type of virtual gift presented by the associated user to the host user. By associating the type of virtual gift with the user's permission level, corresponding permissions can be given according to the user's contribution, thus optimizing the collaborative control effect of the game.

[0131] Specifically, virtual gifts of each gift type usually correspond to different values and weights. The server presets a mapping relationship that corresponds different types of virtual gifts to different permission levels. For example, some high-value virtual gifts may correspond to higher permission levels, while ordinary gifts correspond to lower permission levels. When an associated user presents a virtual gift to the host user, the server queries the preset mapping relationship according to the type of the gift to determine the permission level that the user should obtain.

[0132] In some embodiments, the server can adopt an accumulative method to accumulate the values of all virtual gifts presented by the user to determine the user's permission level. For example, if a user presents multiple different types of virtual gifts, the server will accumulate the values of these gifts and then determine the user's permission level according to the accumulated total value. This accumulative method can more comprehensively reflect the user's contribution and ensure that the setting of the permission level is more reasonable.

[0133] Accordingly, the live streaming server can dynamically adjust the user's permission level according to the user's contribution (reflected by the type and value of virtual gifts), thus better balancing the interactivity and smoothness of the game. This method not only improves the flexibility and adaptability of the game, but also provides a richer and fairer gaming experience for users.

[0134] Step S3130: When the barrage instruction submitted by the associated user is recognized as an organization instruction, parse the barrage instruction into an instruction text, merge it with the collaborative task prompt information corresponding to the set permission level to form a text content, and synthesize the text content into the live image of the cloud game;

[0135] When the barrage instruction submitted by the associated user is recognized as an organization instruction, the server first parses the instruction, extracts its core content and converts it into an instruction text. This text content indicates the specific requirements of the game collaborative task, such as attacking a certain target, supporting a certain teammate or completing a certain specific action, etc. At the same time, the server generates corresponding collaborative task prompt information according to the permission level of the associated user. These prompt information can be the description of the task importance, the suggestions for executing the task or the call to other users, etc., aiming to enhance the guidance and interactivity of the collaborative task.

[0136] Subsequently, the server merges the instruction text with the collaborative task prompt information to form a complete text content. This merged text content not only contains the core intention of the organization instruction, but also incorporates additional information based on the user's permission level, enabling other users to more clearly understand the importance and execution priority of the instruction. Finally, the server embeds this text content into the live image of the cloud game through image synthesis technology, ensuring that all associated users can see the organization instruction and its prompt information in real time when watching the live broadcast.

[0137] There can be various variants in the specific implementation of this processing method. For example, in some embodiments, the server can format the instruction text, such as bolding, changing the color or adding special marks, etc., to highlight the organization instruction. At the same time, the collaborative task prompt information can be presented in the form of a pop-up box, a scroll bar or other visual elements to distinguish it from other live broadcast content. In addition, the server can also dynamically adjust the display mode of the instruction text and the prompt information according to the specific scenario and rules of the game to ensure its readability and effectiveness in different situations.

[0138] By parsing the organization instruction into text, merging it with the collaborative task prompt information and synthesizing it into the live image, the live broadcast server can not only effectively convey the intention of commanding users, but also enhance the influence of the instruction according to the user's permission level, thus optimizing the collaborative control effect of the game. This process not only improves the interactivity and participation of the game, but also provides a more intuitive and orderly game environment for users.

[0139] Step S3140: Broadcast a broadcast message corresponding to the virtual gift giving request to the live broadcast room where the associated user is located, so as to display corresponding notification special effects in the live broadcast room.

[0140] After completing the above process, the live server broadcasts a broadcast message corresponding to the virtual gift giving request to the live room where the associated user is located, so as to display the corresponding notification special effects in the live room. Specifically, when the associated user submits a virtual gift giving request and completes the transfer operation, the server will generate a broadcast message, which contains relevant information about the virtual gift giving, such as the giver, the receiver, the gift type and quantity, etc. Subsequently, the server sends this broadcast message to the live room where the associated user is located to ensure that all users in this live room can receive this notification.

[0141] By broadcasting the virtual gift giving request and displaying the notification special effects, the live server can not only notify other users of the behavior of the associated user in real time, but also enhance the interactive atmosphere and user participation in the live room through visual special effects. This process not only improves the user interaction, but also creates a more active and interesting environment for the live room, further enhancing the user experience.

[0142] In this embodiment, by combining the virtual gift giving with the permission level of the game collaboration task, the interactivity and collaboration efficiency of the bullet screen cloud game are significantly improved. First, by allowing users to give virtual gifts to compete for a higher permission level, a clear and highly motivating participation mechanism is provided for users. This mechanism not only enhances the interactivity between users, but also provides a dynamic basis for permission allocation for the execution of game collaboration tasks, enabling the contribution of users to be directly reflected in their influence on the game process. Second, by broadcasting the virtual gift giving message and displaying the notification special effects, other users can be notified of this behavior in real time, further enhancing the interactive atmosphere and user participation in the live room. This real-time feedback mechanism enables users to more intuitively feel the impact of their actions, while also motivating other users to actively participate in the interaction. In addition, by parsing the organization instruction and generating the collaboration task prompt information in combination with the permission level, the intention of guiding users can be more effectively conveyed, enhancing the authority and guidance of the instruction, thereby optimizing the collaborative control effect of the game. Overall, these steps work together to not only improve the flexibility and adaptability of the game, but also provide users with a richer and fairer game experience, significantly enhancing the overall performance and user experience of the bullet screen cloud game.

[0143] Based on any embodiment of the method of the present application, applying the group instruction to the game process of the cloud game to present the action effect in the live image includes:

[0144] Step S3410, call a preset interface to send the group instruction to the game server of the cloud game, so that the game server applies the group instruction to act on the game process of the cloud game, so that the live image presents the corresponding action effect;

[0145] The live server sends the group instruction to the game server of the cloud game by calling a preset interface, and this process is a key step in realizing the impact of the group instruction on the game process. The preset interface is a communication bridge between the live server and the game server, which allows the live server to transmit the integrated group instruction or the unintegrated barrage instruction to the game server, so that the game server can perform corresponding operations on the game process according to these instructions.

[0146] After receiving the group instruction, the game server will parse the instruction content and identify key information such as the game operation type, overall degree data, and game operation target. Subsequently, the game server operates on the objects or scenes in the game according to this information, thereby changing the current state of the game. For example, if the group instruction is "concentrate on attacking character A", the game server will perform corresponding attack operations on character A according to data such as the overall attack intensity and duration in the instruction, and update the game state in real time.

[0147] By calling the preset interface to send the group instruction to the game server, the live server can not only achieve real-time control of the game process, but also ensure that the update of the game state can be timely fed back to all associated users through the live image. This process not only improves the interactivity and real-time nature of the game, but also provides users with a more smooth and orderly game experience.

[0148] Step S3420: Receive the action effect data returned by the game server after executing the group instruction, and construct corresponding action special effects according to the action effect data;

[0149] To ensure the accurate execution and feedback of the instruction, the game server will return action effect data after executing the group instruction. These data include the result of the operation (such as the state change of the target object, whether the task is completed, etc.) and the visual effects generated during the operation (such as attack animations, special effects, etc.). After receiving these action effect data, the live server will construct corresponding action special effects according to the data content and synthesize these special effects into the live image, so that all associated users can see the effect of the group instruction in real time.

[0150] Specifically, the action effect data can include the result of the operation and the visual effects generated during the operation. The operation result refers to the specific impact of the group instruction on the game state, such as the change in the health value of the target object, the completion of the task, etc.; the visual effect refers to the animations, special effects, etc. presented in the game scene, such as attack animations, explosion effects, character state changes, etc. After executing the group instruction, the game server will generate these action effect data and send them back to the live server.

[0151] After receiving the action effect data, the live server constructs corresponding action special effects according to the data content. This process can be achieved through image processing technology. For example, the visual effects in the action effect data are rendered as image layers and synthesized with the game live stream. For example, if the group instruction is "focus on attacking character A", the action effect data returned by the game server may include the reduction of character A's health value and the attack animation. The live server constructs corresponding attack special effects based on this data and synthesizes them into the live image, so that all associated users can see the effect of character A being attacked in real time.

[0152] By receiving the action effect data returned by the game server and constructing corresponding action special effects, the live server can visually present the effect of the group instruction to all associated users.

[0153] Step S3430: Synthesize the action special effects into the live image.

[0154] The live server synthesizes the constructed action special effects into the live image of the cloud game to ensure that all associated users can view the effect of the group instruction in real time. This process can be achieved through image processing technology. Specifically, the live server renders the action special effects as image layers and synthesizes them with the game live stream.

[0155] The synthesis of action special effects can consider multiple aspects. First, the display position of the action special effects can be matched with the target object or operation area in the game scene. For example, if the group instruction is "focus on attacking character A", then the attack special effects need to be accurately displayed at the position of character A. Second, the style and manifestation form of the action special effects need to be consistent with the theme and style of the game. For example, the attack special effects can be flames, lightning, or other visual effects that match the game settings.

[0156] In some embodiments, the live server can adopt real-time image processing technology to ensure that the synthesis of action special effects is completed within milliseconds to meet the real-time requirements of the game. For example, after the game server returns the action effect data, the live server can immediately render the action special effects as image layers and overlay them with the live image. In addition, the live server can also dynamically adjust the display method of the action special effects according to the specific scene and rules of the game. For example, in fast-paced battle scenes, the action special effects can be more concise and fast; while in scenes with stronger strategy, the action special effects can be more detailed and complex.

[0157] By synthesizing the action special effects into the live image, the live server can not only visually present the effect of the group instruction to all associated users, but also ensure the real-time nature and interactivity of the game.

[0158] In this embodiment, by applying the group instruction to the game process of the cloud game and visually presenting the effect of the application to the live image, the real-time performance and interactivity of the bullet screen cloud game are significantly improved. First, by calling a preset interface to send the group instruction to the game server, efficient control of the game process is achieved, ensuring that the instruction can quickly act on the game state, thereby providing users with a smooth gaming experience. Second, by receiving the action effect data returned by the game server and constructing corresponding action special effects, the live server can visually present the effect of the group instruction to all associated users, enhancing the visual impact of the game and the sense of user participation. Finally, by synthesizing the action special effects into the live image, the live server ensures that all users can view the changes in the game in real time, further optimizing the interactivity and real-time performance of the game. This mechanism not only improves the smoothness of the game but also enhances the sense of cooperation and competition among users, bringing a more rich and vivid gaming experience to the bullet screen cloud game.

[0159] Please refer to Figure 3 , a collaborative control device for a bullet screen cloud game provided according to one aspect of the present application includes an instruction receiving module 3100, a task determination module 3200, an instruction integration module 3300, and an instruction application module 3400. Among them, the instruction receiving module 3100 is configured to receive bullet screen instructions from multiple associated users of the cloud game and synthesize the bullet screen instructions into the live image of the cloud game in the form of text content; the task determination module 3200 is configured to, when detecting an organization instruction sent by a commanding user among the associated users from the bullet screen instructions, determine the current game collaboration task according to the organization instruction; the instruction integration module 3300 is configured to perform the same-type merging of the bullet screen instructions sent by each associated user within the life cycle corresponding to the game collaboration task and adapted to the game collaboration task, and construct them into a group instruction; the instruction application module 3400 is configured to apply the group instruction to the game process of the cloud game to present the effect of the application to the live image.

[0160] Based on any embodiment of the device of the present application, the task determination module 3200 includes: an instruction parsing module configured to parse the organization instruction and determine its corresponding text content; an image acquisition module configured to acquire a plurality of consecutive image frames including the image frame corresponding to the time stamp of the organization instruction in the live image of the cloud game according to the time stamp of the organization instruction; and an inference determination module configured to perform task recognition according to the text content and the plurality of consecutive image frames and determine the game collaboration task corresponding to the organization instruction.

[0161] Based on any embodiment of the device in the present application, prior to the instruction integration module 3300, the device further includes: an instruction parsing module configured to parse each barrage instruction of the multiple associated users generated during the life cycle and extract its text content; a willingness determination module configured to perform task-oriented recognition on each text content to determine the collaborative willingness type, so as to determine whether the corresponding barrage instruction is suitable for the game collaboration task; an independent execution module configured to directly apply the corresponding barrage instruction to the game process of the cloud game alone when the collaborative willingness type belongs to the independent execution type; a collaborative execution module configured to determine that the corresponding barrage instruction is suitable for the game collaboration task when the collaborative willingness type belongs to the non-independent execution type, and is used to construct a group instruction.

[0162] Based on any embodiment of the device in the present application, the instruction integration module 3300 includes: an instruction decomposition module configured to determine the game operation type and action degree data of the corresponding associated user according to the text content of the barrage instruction suitable for the game collaboration task, where the game collaboration task includes a game operation target and a game operation type; a classification processing module configured to classify the barrage instructions of each associated user according to the game operation type, and accumulate the action degree data of each barrage instruction in each category to obtain the overall degree data; a classification construction module configured to construct a corresponding group instruction for each category of barrage instructions, and the group instruction includes the game operation type corresponding to the category of barrage instructions, the overall degree data, and the game operation target specified in the game collaboration task.

[0163] Based on any embodiment of the device in the present application, prior to the instruction integration module, the device further includes: a cycle setting module configured to apply a preset strategy to determine the duration corresponding to the life cycle of the latest organization instruction according to the cumulative consideration of the virtual gifts presented by the commanding user to the host user; a command preemption module configured to detect whether the permission level of the associated user who issues the latest organization instruction exceeds the permission level of the associated user to whom the currently applied organization instruction belongs. When it exceeds, end the currently applied organization instruction and send a notification message to each associated user, and enable the latest organization instruction to make its life cycle take effect.

[0164] Based on any embodiment of the device in the present application, the instruction receiving module 3100 includes: a request execution module configured to receive a virtual gift giving request submitted by any associated user, transfer the virtual gift carried by the request from the personal account of the associated user to the personal account of the host user; a permission configuration module configured to set the permission level of the associated user according to the gift type of the virtual gift given by the associated user to the host user; an image synthesis module configured to, when the barrage instruction submitted by the associated user is recognized as an organization instruction, parse the barrage instruction into instruction text, merge it with the collaborative task prompt information corresponding to the set permission level to form text content, and synthesize the text content into the live image of the cloud game; a message broadcast module configured to broadcast a broadcast message corresponding to the virtual gift giving request to the live room where the associated user is located, so as to display a corresponding notification special effect in the live room.

[0165] Based on any embodiment of the device in the present application, the instruction application module 3400 includes: a transmission execution module configured to call a preset interface to send the group instruction to the game server of the cloud game, so that the game server applies the group instruction to act on the game process of the cloud game, so that the live image presents a corresponding action effect; a special effect construction module configured to receive the action effect data returned by the game server after executing the group instruction, and construct a corresponding action special effect according to the action effect data; a special effect presentation module configured to synthesize the action special effect into the live image.

[0166] Another embodiment of the present application further provides a collaborative control device for barrage cloud games. As Figure 4 shown, it is a schematic internal structure diagram of the collaborative control device for barrage cloud games. The collaborative control device for barrage cloud games includes a processor, a computer-readable storage medium, a memory, and a network interface connected through a system bus. Among them, the computer-readable non-volatile storage medium of the collaborative control device for barrage cloud games stores an operating system, a database, and computer-readable instructions. Information sequences can be stored in the database. When the computer-readable instructions are executed by the processor, the processor can implement a collaborative control method for barrage cloud games.

[0167] The processor of the collaborative control device for barrage cloud games is used to provide computing and control capabilities to support the operation of the entire collaborative control device for barrage cloud games. Computer-readable instructions can be stored in the memory of the collaborative control device for barrage cloud games. When the computer-readable instructions are executed by the processor, the processor can execute the collaborative control method for barrage cloud games of the present application. The network interface of the collaborative control device for barrage cloud games is used to connect and communicate with the terminal.

[0168] Those skilled in the art can understand, Figure 4The structure shown is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the barrage cloud game collaborative control device to which the solution of this application is applied. Specifically, the barrage cloud game collaborative control device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0169] In this embodiment, the processor is used to execute Figure 3 the specific functions of each module in. The memory stores the program codes and various types of data required to execute the above-mentioned modules or sub-modules. The network interface is used to implement data transmission between the user terminal or the server. In this embodiment, the non-volatile readable storage medium stores the program codes and data required to execute all modules in the barrage cloud game collaborative control device of this application, and the server can call the program codes and data of the server to execute the functions of all modules.

[0170] This application also provides a non-volatile readable storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors are caused to execute the steps of the barrage cloud game collaborative control method according to any embodiment of this application.

[0171] This application also provides a computer program product, including computer programs / instructions, which when executed by one or more processors, implement the steps of the method according to any embodiment of this application.

[0172] Those of ordinary skill in the art can understand that to implement all or part of the processes in the methods of the above embodiments of this application, it can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile readable storage medium. When the program is executed, it may include the processes of the embodiments of the above methods. Among them, the aforementioned storage medium may be a computer-readable storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0173] In summary, the application of the technical solution of this application has significantly improved the interactivity and user experience of barrage cloud games in the network live broadcast scenario. By introducing the command user and organization instructions, efficient collaboration among users has been achieved, making the game process more orderly and smooth. This innovative collaboration mode not only enhances the interaction between users, but also brings more gameplay and fun to the game, expanding the business model. At the same time, by combining the barrage instructions of multiple users into group instructions, the processing burden on the server is reduced, the response efficiency is improved, the game latency is reduced, and users can participate in the game more smoothly. In addition, this efficient collaboration mode also enhances the user's sense of participation and satisfaction, further optimizing the user experience.

Claims

1. A collaborative control method for a bullet screen cloud game, characterized in that: include: Receiving bullet screen instructions from multiple associated users of the cloud game, and synthesizing the bullet screen instructions into the live broadcast image of the cloud game in the form of text content; When an organization instruction sent by a commanding user among the associated users is detected from the barrage instruction, determining a current game collaboration task according to the organization instruction; Merge the same type of bullet screen instructions sent by each associated user within the corresponding life cycle of the game collaborative task and adapted to the game collaborative task to construct a group instruction; The group command is applied to the game process of the cloud game to present the effect in the live broadcast image.

2. The collaborative control method of the bullet screen cloud game according to claim 1 is characterized in that: Determine the current game collaboration task according to the organization instruction, including: Parsing the organization instruction to determine the corresponding text content; Acquire, according to the timestamp of the organization instruction, a plurality of continuous image frames in the live broadcast image of the cloud game including an image frame corresponding to the timestamp; Task identification is performed based on the text content and the multiple continuous image frames to determine the game collaboration task corresponding to the organization instruction.

3. The collaborative control method of the bullet screen cloud game according to claim 1 is characterized in that: The bullet screen instructions sent by each associated user within the corresponding life cycle of the game collaborative task and adapted to the game collaborative task are merged into a group instruction of the same type, including: Parsing each bullet screen instruction of the plurality of associated users generated in the life cycle to extract the text content thereof; Performing task orientation identification on each of the text contents to determine the type of collaborative willingness, so as to determine whether the corresponding bullet screen instruction is suitable for the game collaborative task; When the coordination intention type belongs to the autonomous execution type, the corresponding bullet screen instruction is directly applied to the game process of the cloud game; When the collaboration intention type belongs to the non-autonomous execution type, the corresponding bullet screen instructions are determined to adapt to the game collaboration task for constructing group instructions.

4. The collaborative control method of the bullet screen cloud game according to claim 1, characterized in that: The bullet screen instructions adapted to the game collaborative task sent by each associated user within the corresponding life cycle of the game collaborative task are merged into a group instruction of the same type, including: Determining the game operation type and action degree data of the corresponding associated user according to the text content of the bullet screen instruction adapted to the game collaborative task, wherein the game collaborative task includes a game operation target and a game operation type; Classify the bullet screen commands of each associated user according to the game operation type, and accumulate the action degree data of each bullet screen command in each type of bullet screen command to obtain the overall degree data; Corresponding group instructions are constructed for each type of barrage instruction, and the group instruction includes the game operation type corresponding to the type of barrage instruction, the overall degree data, and the game operation target specified in the game collaborative task.

5. The collaborative control method of the bullet screen cloud game according to claim 1, characterized in that: Before merging the same type of bullet screen instructions sent by each associated user within the corresponding life cycle of the game collaborative task and adapted to the game collaborative task to construct a group instruction, the method includes: Applying a preset strategy, determining the duration corresponding to the life cycle of the latest organization instruction according to the accumulated consideration of the virtual gifts given by the commanding user to the anchor user; Check whether the authority level of the associated user who issued the latest organization instruction exceeds the authority level of the associated user to which the currently applied organization instruction belongs. If it exceeds, terminate the currently applied organization instruction and send a notification message to each associated user, enable the latest organization instruction and make its life cycle take effect.

6. The collaborative control method of the bullet screen cloud game according to claim 5 is characterized in that: Receiving bullet screen instructions from multiple associated users of the cloud game, and synthesizing the bullet screen instructions into the live broadcast image of the cloud game in the form of text content, including: Receive a virtual gift donation request submitted by any associated user, and transfer the virtual gift carried in the request from the associated user's personal account to the anchor user's personal account; According to the gift type of the virtual gift presented by the associated user to the anchor user, the permission level of the associated user is correspondingly set; When the bullet screen instruction submitted by the associated user is identified as an organization instruction, the bullet screen instruction is parsed into an instruction text, and the collaborative task prompt information corresponding to the set permission level is merged to form text content, and the text content is synthesized into the live broadcast image of the cloud game; The broadcast message corresponding to the virtual gift donation request is broadcast to the live broadcast room where the associated user is located, so as to display the corresponding notification effect in the live broadcast room.

7. The collaborative control method of the bullet screen cloud game according to any one of claims 1 to 6, characterized in that: Applying the group command to the game process of the cloud game to present the effect to the live broadcast image includes: Calling a preset interface to send the group instruction to the game server of the cloud game, so that the game server applies the group instruction to act on the game process of the cloud game, so that the live broadcast image presents a corresponding effect; Receiving action effect data returned by the game server after executing the group command, and constructing corresponding action special effects according to the action effect data; The action special effects are synthesized into the live broadcast image.

8. A collaborative control device for a bullet screen cloud game, characterized in that: include: An instruction receiving module, configured to receive bullet screen instructions from multiple associated users of the cloud game, and synthesize the bullet screen instructions into the live broadcast image of the cloud game in the form of text content; A task determination module, configured to determine a current game collaboration task according to the organization instruction when an organization instruction sent by a commanding user among the associated users is detected from the bullet screen instruction; The instruction integration module is configured to merge the bullet screen instructions of the same type sent by each associated user within the corresponding life cycle of the game collaborative task and adapted to the game collaborative task to construct a group instruction; The instruction application module is configured to apply the group instruction to the game process of the cloud game to present the effect in the live broadcast image.

9. A bullet screen cloud game collaborative control device, comprising a central processing unit and a memory, characterized in that: The central processing unit is used to call and run the computer program stored in the memory to execute the steps of the method according to any one of claims 1 to 7.

10. A non-volatile readable storage medium, characterized in that: It stores a computer program implemented according to the method described in any one of claims 1 to 7 in the form of computer-readable instructions, and when the computer program is called and executed by a computer, the steps included in the corresponding method are executed.

Images