A method, apparatus, device, and storage medium for creating live streaming tasks.

Abstract

This application discloses a method, apparatus, device, and storage medium for creating live streaming tasks, relating to the field of live streaming creation. The method includes: acquiring a live streaming task creation request, and generating live streaming data based on a preset formatted template, the live streaming task creation request, and the live streaming task creation time corresponding to the request; inputting the live streaming data into a pre-trained large language model to determine the live streaming task type corresponding to the live streaming data and the information of each live streaming decomposition task corresponding to the live streaming task type; and outputting corresponding standardized task information based on the live streaming data and the information of each live streaming decomposition task; and creating corresponding system live streaming tasks locally based on the standardized task information. This application analyzes live streaming data using a large language model, automatically outputs standardized task information related to live streaming task creation, and then creates corresponding system live streaming tasks locally, achieving unified creation of live streaming-related tasks.

Description

Technical Field

[0001] This invention relates to the field of live streaming creation, and in particular to a method, apparatus, device, and storage medium for creating live streaming tasks. Background Technology

[0002] The live streaming industry is currently experiencing rapid growth and has expanded to numerous sectors, including e-commerce, entertainment, gaming, and sports, with a continuously increasing user base. However, existing live streaming systems typically use traditional input methods such as lists and forms to create live streaming tasks. This means that each sub-task within a single live streaming task is usually created one by one, which is inefficient. Therefore, how to achieve unified creation of all sub-tasks within a live streaming task is a problem that needs to be solved. Summary of the Invention

[0003] In view of this, the purpose of this invention is to provide a method, apparatus, device, and storage medium for creating live streaming tasks. This method can analyze live streaming data using a large language model, automatically output standardized task information related to live streaming task creation, and then create corresponding live streaming tasks for each system locally, achieving unified creation of live streaming-related tasks. The specific solution is as follows:

[0004] Firstly, this application provides a method for creating live streaming tasks, applied to a pre-defined cloud system, including:

[0005] Obtain a live streaming task creation request, and generate live streaming data based on a preset format template, the live streaming task creation request, and the live streaming task creation time corresponding to the live streaming task creation request;

[0006] The live streaming data is input into a pre-trained large language model to determine the live streaming task type corresponding to the live streaming data and the information of each live streaming decomposition task corresponding to the live streaming task type. Based on the live streaming data and the information of each live streaming decomposition task, the corresponding standardized task information is output.

[0007] Based on the standardized task information described above, corresponding live streaming tasks for each system are created locally.

[0008] Optionally, obtaining the live streaming task creation request includes:

[0009] The input for creating a live streaming task is obtained through a dialog window in the application layer.

[0010] Optionally, generating live data based on a preset format template, the live task creation request, and the live task creation time corresponding to the live task creation request includes:

[0011] Obtain the live stream task creation time corresponding to the live stream task creation request;

[0012] Based on a preset format template and the creation time of the live streaming task, prefixes and suffixes are added to the live streaming task creation request to generate live streaming data.

[0013] Optionally, before inputting the live data into the pre-trained large language model, the method further includes:

[0014] Retrieve historical live stream association data from the local database, and obtain the historical live stream time and historical live stream task creation time corresponding to the historical live stream task from the historical live stream association data;

[0015] Determine the historical time difference between the historical live broadcast time and the historical live broadcast task creation time, and determine the live broadcast task type based on the historical time difference. Then, determine the information of each live broadcast decomposition task corresponding to the live broadcast task type from the historical live broadcast association data.

[0016] The historical live stream association data, the live stream task type, and the information of each live stream decomposition task corresponding to the live stream task type are formatted using a preset model training template to obtain formatted live stream data. The formatted live stream data is then used to train the initial large language model to obtain a trained large language model.

[0017] Optionally, the method further includes:

[0018] If the pre-trained large language model does not output the standardized task information, then the live task type corresponding to the live data is output by the pre-trained large language model, and the live data is regenerated based on the preset format template, the live task creation request, and the live task creation time corresponding to the live task creation request. Then, the process jumps to the step of inputting the live data into the pre-trained large language model.

[0019] Optionally, determining the live streaming task type corresponding to the live streaming data and the information of each live streaming decomposition task corresponding to the live streaming task type through the pre-trained large language model includes:

[0020] The live streaming time and the live streaming task creation time contained in the live streaming data are obtained through the pre-trained large language model, and the live streaming time difference between the live streaming time and the live streaming task creation time is calculated.

[0021] The pre-trained large language model determines the live streaming task type corresponding to the live streaming data based on the live streaming time difference, and determines the information of each live streaming decomposition task corresponding to the live streaming task type.

[0022] Optionally, the step of outputting corresponding standardized task information based on the live streaming data and the information of each live streaming decomposed task includes:

[0023] The live streamer information contained in the live stream data is obtained through the pre-trained large language model, and the live stream group corresponding to the live streamer information is determined; and the correspondence between each live streamer in the live stream group and each of the live stream decomposition task information is determined.

[0024] The pre-trained large language model outputs corresponding standardized task information based on the live data, the information of each live decomposition task, and the corresponding relationship.

[0025] Secondly, this application provides a live streaming task creation device, applied to a pre-set cloud system, comprising:

[0026] The live data generation module is used to obtain a live task creation request and generate live data based on a preset format template, the live task creation request, and the live task creation time corresponding to the live task creation request.

[0027] The task information output module is used to input the live data into a pre-trained large language model, so as to determine the live task type corresponding to the live data and the live decomposition task information corresponding to the live task type through the pre-trained large language model, and output the corresponding standardized task information based on the live data and the live decomposition task information.

[0028] The live streaming task creation module is used to create corresponding live streaming tasks for each system locally based on the standardized task information described above.

[0029] Thirdly, this application provides an electronic device, comprising:

[0030] Memory, used to store computer programs;

[0031] A processor for executing the computer program to implement the aforementioned live streaming task creation method.

[0032] Fourthly, this application provides a computer-readable storage medium for storing a computer program, which, when executed by a processor, implements the aforementioned live streaming task creation method.

[0033] In this application, a live streaming task creation request is obtained, and live streaming data is generated based on a preset format template, the live streaming task creation request, and the live streaming task creation time corresponding to the request. The live streaming data is input into a pre-trained large language model to determine the live streaming task type and the corresponding live streaming decomposition task information. Based on the live streaming data and the live streaming decomposition task information, corresponding standardized task information is output. Based on the standardized task information, corresponding system live streaming tasks are created locally. Therefore, this application analyzes live streaming data using a pre-trained large language model, first determining the live streaming task type, then automatically outputting standardized task information related to live streaming task creation based on the live streaming task decomposition information, and finally creating corresponding system live streaming tasks locally on a preset cloud system based on the standardized task information. This achieves unified creation of live streaming-related tasks, improves the automation level of live streaming task creation, and increases the efficiency of live streaming task creation. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0035] Figure 1 This application discloses a flowchart of a method for creating a live streaming task.

[0036] Figure 2 This application discloses a flowchart of a large language model training process.

[0037] Figure 3 A schematic diagram of a live streaming task creation device disclosed in this application;

[0038] Figure 4 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] Existing live streaming systems typically use traditional input methods such as lists and forms to create live streaming tasks. This often involves creating each sub-task within a single task individually, resulting in low efficiency. Therefore, this application provides a live streaming task creation method that analyzes live streaming data using a large language model, automatically outputs standardized task information related to live streaming task creation, and then creates the corresponding system live streaming tasks locally, achieving unified creation of live streaming-related tasks.

[0041] See Figure 1 As shown, this embodiment of the invention discloses a method for creating live streaming tasks, applied to a preset cloud system, including:

[0042] Step S11: Obtain the live streaming task creation request, and generate live streaming data based on the preset format template, the live streaming task creation request, and the live streaming task creation time corresponding to the live streaming task creation request.

[0043] In this embodiment, obtaining the live streaming task creation request may include acquiring the input live streaming task creation request through a dialog window in the application layer. Specifically, by adding a dialog window in the application layer, the user can input the live streaming task creation request in the dialog window, and the cloud system can then obtain the input live streaming task creation request. The live streaming task creation request can be in text or voice format, etc., and is not limited thereto; furthermore, the live streaming task creation request includes, but is not limited to, information about the live streamers, live streaming time, and live streaming actions.

[0044] In this embodiment, generating live data based on a preset format template, the live task creation request, and the live task creation time corresponding to the live task creation request may include obtaining the live task creation time corresponding to the live task creation request; and adding prefixes and suffixes to the live task creation request based on the preset format template and the live task creation time to generate live data. It can be understood that the live task creation time corresponding to the live task creation request is first obtained, then the preset format template is used as a prefix, and the live task creation time is used as a suffix, which are added to the beginning and end of the live task creation request to obtain the live data. For example, the default format template is "Task Name|Responsible Person|Live Broadcast Time|Creation Time|Task Objective". The live broadcast task creation request is "I want to hold a live broadcast on April 30th, please help me create and break down the task". The corresponding live broadcast task creation time is "Wednesday, April 11, 2023". Adding prefixes and suffixes to the live broadcast task creation request will generate the following live broadcast data: "Please organize and format the following requirements according to Task Name|Responsible Person|Live Broadcast Time|Creation Time|Task Objective: I want to hold a live broadcast on April 30th, please help me create and break down the task. The live broadcast task creation time is Wednesday, April 11, 2023".

[0045] Step S12: Input the live data into a pre-trained large language model to determine the live task type corresponding to the live data and the information of each live decomposition task corresponding to the live task type through the pre-trained large language model, and output the corresponding standardized task information based on the live data and each of the live decomposition task information.

[0046] In this embodiment, the generated live streaming data is input into a pre-trained large language model to obtain the live streaming time and the live streaming task creation time contained in the live streaming data, and to calculate the difference between the live streaming time and the live streaming task creation time. Then, the pre-trained large language model determines the live streaming task type corresponding to the live streaming data based on the live streaming time difference, and determines the information of each live streaming decomposition task corresponding to the live streaming task type. It can be understood that the pre-trained large language model first obtains the live streaming time "April 30th" and the live streaming task creation time "April 11th" from the live streaming data, calculates the difference between the live streaming time and the live streaming task creation time "19 days". Since different live streaming task types correspond to different difference ranges, the live streaming task type corresponding to the live streaming data can be determined based on the difference. Furthermore, a live streaming task usually contains multiple decomposition tasks, and different live streaming task types correspond to different live streaming decomposition task information; therefore, the information of each corresponding live streaming decomposition task can be determined based on the live streaming task type.

[0047] In this embodiment, the step of outputting corresponding standardized task information based on the live streaming data and each of the live streaming decomposition task information may include obtaining the live streamer information contained in the live streaming data through the pre-trained large language model, and determining the live streaming group corresponding to the live streamer information; determining the correspondence between each live streamer in the live streaming group and each of the live streaming decomposition task information; and outputting corresponding standardized task information based on the live streaming data, each of the live streaming decomposition task information, and the correspondence through the pre-trained large language model. It is understood that the pre-trained large language model obtains the live streamer information corresponding to "me" from the live streaming data. Each live streamer information will have a corresponding live streaming group, and each live streamer in the live streaming group corresponds to different live streaming decomposition task information. Therefore, it is necessary to output corresponding standardized task information based on the correspondence between each live streamer in the live streaming group and each of the live streaming decomposition task information, each of the live streaming decomposition task information, and the preset format template, live streaming time, and live streaming task creation time contained in the live streaming data. For example, the live stream task breakdown information includes live stream plan creation and live stream planning preparation. The live stream plan creation corresponds to the live stream personnel A, and the live stream planning preparation corresponds to the live stream personnel B. The preset format template is Task Name | Person in Charge | Live Stream Time | Creation Time | Task Objective. The generated standardized task information is as follows:

[0048] Task Name 1: Live Streaming Plan Creation;

[0049] Person in charge: A;

[0050] Live broadcast time: April 30th;

[0051] Creation time: April 11;

[0052] Task objective: Create a live streaming plan in the preset cloud system directory xxx;

[0053] Task Name Two: Live Stream Planning and Preparation;

[0054] Person in charge: B;

[0055] Completion date: April 30th;

[0056] Creation time: April 11;

[0057] Task objective: Complete the planning and preparation for the live broadcast.

[0058] In this embodiment, if the pre-trained large language model does not output the standardized task information, the live task type corresponding to the live data output by the pre-trained large language model is used to regenerate the live data based on the preset format template, the live task creation request, and the live task creation time corresponding to the live task creation request. Then, the process jumps to the step of inputting the live data into the pre-trained large language model. It can be understood that if the pre-trained large language model does not output standardized task information, it will output the live task type corresponding to the live data to the preset cloud system, which includes the remote cloud system. After receiving the output live task type, the preset cloud system regenerates the live data using the live task type, the preset format template, the live task creation request, and the live task creation time. Specifically, the live task type and the preset format template are used as prefixes, and the live task creation time is used as a suffix. Prefix and suffix additions are performed on the live task creation request to regenerate the live data, and then the process jumps to step S12, where the live data is input into the pre-trained large language model.

[0059] Step S13: Create corresponding live streaming tasks for each system locally based on the standardized task information.

[0060] In this embodiment, the preset cloud system automatically creates corresponding live streaming tasks for each system locally using the standardized task information output by the pre-trained large language model. Furthermore, the preset cloud system identifies the standardized task information and then automatically converts it into corresponding live streaming tasks, and distributes each live streaming task to the corresponding live streaming personnel.

[0061] Therefore, this application analyzes live streaming data using a pre-trained large language model, first determining the live streaming task type corresponding to the live streaming data, then automatically outputting standardized task information related to the creation of live streaming tasks based on the decomposition information of each live streaming task type, and finally creating corresponding system live streaming tasks locally in a preset cloud system based on the standardized task information, thereby achieving unified creation of live streaming related tasks, improving the automation level of live streaming task creation, and improving the efficiency of live streaming task creation.

[0062] As described in the previous embodiment, this application describes the process of analyzing live streaming data using a pre-trained large language model. Next, this application will elaborate on how to train the large language model. See [link to previous document]. Figure 2 As shown, this embodiment of the invention discloses a process for training a large language model, including:

[0063] Step S21: Obtain historical live streaming association data from the local database, and obtain the historical live streaming time and historical live streaming task creation time corresponding to the historical live streaming task in the historical live streaming association data.

[0064] In this embodiment, the cloud system pre-determines that all historical live streaming associated data are retrieved from a local database. This historical live streaming associated data includes, but is not limited to, anchor information, live streamer information, live stream team information, live stream task information, live stream time, and live stream task creation time. Furthermore, each historical live stream task, along with its corresponding historical live stream time and historical live stream task creation time, are retrieved from the historical live stream associated data.

[0065] Step S22: Determine the historical time difference between the historical live broadcast time and the historical live broadcast task creation time, determine the live broadcast task type based on the historical time difference, and then determine the information of each live broadcast decomposition task corresponding to the live broadcast task type from the historical live broadcast association data.

[0066] In this embodiment, the historical time difference between the historical live broadcast time and the historical live broadcast task creation time corresponding to each historical live broadcast task is calculated, and the live broadcast task type is determined based on each historical time difference. Further, if the historical time difference is within 7 days, the corresponding live broadcast task type is Class A; if the historical time difference is between 7 and 14 days, the corresponding live broadcast task type is Class B; and if the historical time difference is greater than 14 days, the corresponding live broadcast task type is Class C. After determining the live broadcast task type, the information of each live broadcast decomposition task corresponding to each live broadcast task type is determined from the historical live broadcast association data. For example, the task breakdown information for Category A live streaming tasks includes, but is not limited to, creating a live streaming plan, quickly selecting products, quickly finalizing products, generating promotional materials, reserving a live streaming room, scheduling the streamer's schedule, reviewing the products broadcast, and conducting a live streaming debriefing. The task breakdown information for Category B live streaming tasks includes, but is not limited to, creating a live streaming plan, preparing for live streaming planning, general product selection, general product finalization, generating promotional materials, reserving a live streaming room, scheduling the streamer's schedule, reviewing the products broadcast, and conducting a live streaming debriefing. The task breakdown information for Category C live streaming tasks includes, but is not limited to, creating a live streaming plan, preparing for live streaming planning, selecting products for the event, finalizing the products for the event, generating promotional materials, reserving a live streaming room, scheduling the streamer's schedule, reviewing the products broadcast, and conducting a live streaming debriefing.

[0067] Step S23: Format the historical live broadcast association data, the live broadcast task type, and the information of each live broadcast decomposition task corresponding to the live broadcast task type using a preset model training template to obtain formatted live broadcast data, and use the formatted live broadcast data to train the initial large language model to obtain a trained large language model.

[0068] In this embodiment, after obtaining the live streaming task type and the corresponding live streaming decomposition task information, the historical live streaming associated data, the live streaming task type, and the corresponding live streaming decomposition task information are formatted using a preset model training template to obtain formatted live streaming data. For example, if the preset model training template is "anchor information | live streaming personnel information | live streaming time | live streaming team | category | brand | merchant | live streaming task", then the historical live streaming associated data, the live streaming task type, and the corresponding live streaming decomposition task information are formatted according to the preset model training template, and the obtained formatted live streaming data is input into the initial large language model to train the initial large language model and obtain a trained large language model. The initial large language model can be OpenAI's publicly available GPT (Generative Pre-Trained Transformer). Additionally, a base layer service and middleware service for connecting to the large language model are added to the preset cloud system to encapsulate the large language model's API (Application Programming Interface) and intermediate parsing, respectively.

[0069] Therefore, this application determines the live streaming task type based on the difference between the historical live streaming time and the historical live streaming task creation time corresponding to each historical live streaming task, and determines the information of each live streaming decomposition task corresponding to the live streaming task type from the historical live streaming associated data. Then, it formats the historical live streaming associated data, the live streaming task type, and the information of each live streaming decomposition task corresponding to the live streaming task type according to the preset model training template. This allows the formatted live streaming data to be used to train the initial large language model, so that the pre-trained large language model can be used to directly analyze the user's input live streaming task creation request, determine the corresponding live streaming task type and the information of each live streaming decomposition task corresponding to the live streaming task type, and output each standardized task information based on the information of each live streaming decomposition task. This achieves standardization of live streaming creation-related tasks, improves the automation level of live streaming task creation, and improves the efficiency of live streaming task creation.

[0070] See Figure 3 As shown, this embodiment of the invention discloses a live streaming task creation device, applied to a preset cloud system, comprising:

[0071] The live data generation module 11 is used to obtain a live task creation request and generate live data based on a preset format template, the live task creation request, and the live task creation time corresponding to the live task creation request.

[0072] The task information output module 12 is used to input the live data into a pre-trained large language model, so as to determine the live task type corresponding to the live data and the live decomposition task information corresponding to the live task type through the pre-trained large language model, and output the corresponding standardized task information based on the live data and the live decomposition task information.

[0073] The live streaming task creation module 13 is used to create corresponding live streaming tasks for each system locally based on the standardized task information.

[0074] Therefore, this application analyzes live streaming data using a pre-trained large language model, first determining the live streaming task type corresponding to the live streaming data, then automatically outputting standardized task information related to the creation of live streaming tasks based on the decomposition information of each live streaming task type, and finally creating corresponding system live streaming tasks locally in a preset cloud system based on the standardized task information, thereby achieving unified creation of live streaming related tasks, improving the automation level of live streaming task creation, and improving the efficiency of live streaming task creation.

[0075] In some specific embodiments, the live data generation module 11 may specifically include:

[0076] The request retrieval unit is used to retrieve the input of the live task creation request through the dialog window in the application layer.

[0077] In some specific embodiments, the live data generation module 11 may specifically include:

[0078] The creation time acquisition unit is used to acquire the creation time of the live streaming task corresponding to the live streaming task creation request;

[0079] The prefix and suffix addition unit is used to add prefixes and suffixes to the live task creation request based on a preset format template and the live task creation time to generate live data.

[0080] In some specific embodiments, the live streaming task creation device may further include:

[0081] The historical data acquisition unit is used to acquire historical live streaming associated data from the local database, and to acquire the historical live streaming time and historical live streaming task creation time corresponding to the historical live streaming task in the historical live streaming associated data.

[0082] The task type determination unit is used to determine the historical time difference between the historical live broadcast time and the historical live broadcast task creation time, and to determine the live broadcast task type based on the historical time difference. Then, it determines the information of each live broadcast decomposition task corresponding to the live broadcast task type from the historical live broadcast association data.

[0083] The model training unit is used to format the historical live broadcast association data, the live broadcast task type, and the information of each live broadcast decomposition task corresponding to the live broadcast task type using a preset model training template to obtain formatted live broadcast data, and to train the initial large language model using the formatted live broadcast data to obtain a trained large language model.

[0084] In some specific embodiments, the live streaming task creation device may further include:

[0085] The step jump unit is used to, if the pre-trained large language model does not output the standardized task information, utilize the live task type output by the pre-trained large language model corresponding to the live data, and regenerate the live data based on the preset format template, the live task creation request, and the live task creation time corresponding to the live task creation request, and then jump to the step of inputting the live data into the pre-trained large language model.

[0086] In some specific embodiments, the task information output module 12 may specifically include:

[0087] The difference calculation unit is used to obtain the live streaming time and the live streaming task creation time contained in the live streaming data through the pre-trained large language model, and to calculate the difference between the live streaming time and the live streaming task creation time.

[0088] The decomposition information determination unit is used to determine the live task type corresponding to the live data based on the live time difference using the pre-trained large language model, and to determine the live decomposition task information corresponding to the live task type.

[0089] In some specific embodiments, the task information output module 12 may specifically include:

[0090] The correspondence determination unit is used to obtain the live streamer information contained in the live stream data through the pre-trained large language model, and determine the live stream group corresponding to the live streamer information; and determine the correspondence between each live streamer in the live stream group and each of the live stream decomposition task information.

[0091] The task information output unit is used to output corresponding standardized task information based on the pre-trained large language model, the live data, the live decomposition task information, and the correspondence.

[0092] Furthermore, embodiments of this application also disclose an electronic device, Figure 4This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.

[0093] Figure 4 This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the live streaming task creation method disclosed in any of the foregoing embodiments. Alternatively, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0094] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0095] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0096] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the live streaming task creation method executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include a computer program capable of performing other specific tasks.

[0097] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned live streaming task creation method. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0098] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0099] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0100] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0101] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0102] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for creating live streaming tasks, characterized in that, Applied to pre-built cloud systems, including: Obtain a live streaming task creation request, and generate live streaming data based on a preset format template, the live streaming task creation request, and the live streaming task creation time corresponding to the live streaming task creation request; The live streaming data is input into a pre-trained large language model to determine the live streaming task type corresponding to the live streaming data and the information of each live streaming decomposition task corresponding to the live streaming task type. Based on the live streaming data and the information of each live streaming decomposition task, the corresponding standardized task information is output. Based on the standardized task information described above, create corresponding live streaming tasks for each system locally; The step of determining the live streaming task type corresponding to the live streaming data and the information of each live streaming decomposition task corresponding to the live streaming task type through the pre-trained large language model includes: The live streaming time and the live streaming task creation time contained in the live streaming data are obtained through the pre-trained large language model, and the live streaming time difference between the live streaming time and the live streaming task creation time is calculated. The pre-trained large language model determines the live streaming task type corresponding to the live streaming data based on the live streaming time difference, and determines the information of each live streaming decomposition task corresponding to the live streaming task type.

2. The live streaming task creation method according to claim 1, characterized in that, The process of obtaining the live streaming task creation request includes: The input for creating a live streaming task is obtained through a dialog window in the application layer.

3. The live streaming task creation method according to claim 1, characterized in that, The process of generating live streaming data based on a preset format template, the live streaming task creation request, and the live streaming task creation time corresponding to the live streaming task creation request includes: Obtain the live stream task creation time corresponding to the live stream task creation request; Based on a preset format template and the creation time of the live streaming task, prefixes and suffixes are added to the live streaming task creation request to generate live streaming data.

4. The live streaming task creation method according to claim 1, characterized in that, Before inputting the live data into the pre-trained large language model, the method further includes: Retrieve historical live stream association data from the local database, and obtain the historical live stream time and historical live stream task creation time corresponding to the historical live stream task from the historical live stream association data; Determine the historical time difference between the historical live broadcast time and the historical live broadcast task creation time, and determine the live broadcast task type based on the historical time difference. Then, determine the information of each live broadcast decomposition task corresponding to the live broadcast task type from the historical live broadcast association data. The historical live stream association data, the live stream task type, and the information of each live stream decomposition task corresponding to the live stream task type are formatted using a preset model training template to obtain formatted live stream data. The formatted live stream data is then used to train the initial large language model to obtain a trained large language model.

5. The live streaming task creation method according to claim 1, characterized in that, Also includes: If the pre-trained large language model does not output the standardized task information, then the live task type corresponding to the live data is output by the pre-trained large language model, and the live data is regenerated based on the preset format template, the live task creation request, and the live task creation time corresponding to the live task creation request. Then, the process jumps to the step of inputting the live data into the pre-trained large language model.

6. The method for creating a live streaming task according to any one of claims 1 to 5, characterized in that, The step of outputting standardized task information corresponding to the live streaming data and the information of each live streaming decomposed task includes: The live streamer information contained in the live stream data is obtained through the pre-trained large language model, and the live stream group corresponding to the live streamer information is determined; and the correspondence between each live streamer in the live stream group and each of the live stream decomposition task information is determined. The pre-trained large language model outputs corresponding standardized task information based on the live data, the information of each live decomposition task, and the corresponding relationship.

7. A live streaming task creation device, characterized in that, Applied to pre-built cloud systems, including: The live data generation module is used to obtain a live task creation request and generate live data based on a preset format template, the live task creation request, and the live task creation time corresponding to the live task creation request. The task information output module is used to input the live data into a pre-trained large language model, so as to determine the live task type corresponding to the live data and the live decomposition task information corresponding to the live task type through the pre-trained large language model, and output the corresponding standardized task information based on the live data and the live decomposition task information. The live streaming task creation module is used to create corresponding live streaming tasks for each system locally based on the standardized task information described above. Specifically, the task information output module is used to: obtain the live streaming time and the live streaming task creation time contained in the live streaming data through the pre-trained large language model, and calculate the live streaming time difference between the live streaming time and the live streaming task creation time; determine the live streaming task type corresponding to the live streaming data based on the live streaming time difference through the pre-trained large language model, and determine the live streaming decomposition task information corresponding to the live streaming task type.

8. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the live streaming task creation method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, Used to store a computer program, which, when executed by a processor, implements the live streaming task creation method as described in any one of claims 1 to 6.

Images