Live broadcast interaction data processing method and device, electronic equipment, medium and product

Abstract

The embodiment of the invention discloses a live broadcast interaction data processing method and device, electronic equipment, a medium and a product. The method comprises the following steps: acquiring first explanation data generated in a target live broadcast; determining first interaction information corresponding to the target live broadcast based on the first explanation data and a pre-trained information identification model; wherein the first interaction information comprises first interaction content and a first action identifier; and sending the first interaction information to the target user side, thereby enabling the target user side to execute a live broadcast interaction operation on a live broadcast interface based on the first interaction information in the target live broadcast. According to the technical scheme provided by the embodiment of the invention, interaction in the target live broadcast playing process can be realized, and the interaction process does not need manual operation, so that the accuracy and efficiency of live broadcast interaction are improved, and the waste of human resources and time resources is reduced.

Description

Technical Field

[0001] Embodiments of the present invention relate to the fields of artificial intelligence and e-commerce technology, and in particular, to a live interactive data processing method, device, electronic device, medium and product. Background Art

[0002] With the rapid development of the Internet, live broadcasting is becoming more and more common in our lives. In live broadcasting scenarios, in order to interact with users, operators are usually required to watch the live broadcast, determine the interactive content based on the live broadcast content, and perform interactive operations on the interactive content.

[0003] However, in the process of implementing the present invention, it was found that the prior art has at least the following technical problems: operators manually determine the interactive content and perform interactive operations, which is prone to errors and slow to operate, resulting in untimely interaction and poor operation accuracy; and it requires a lot of human resources and time resources, resulting in a waste of resources. Summary of the invention

[0004] The embodiments of the present invention provide a live interactive data processing method, device, electronic device, medium and product to achieve the purpose of improving the accuracy and efficiency of live interactive and reducing the waste of human and time resources.

[0005] According to one aspect of the present invention, a live interactive data processing method is provided, comprising:

[0006] Obtain the first explanation data generated in the target live broadcast;

[0007] Based on the first explanation data and the pre-trained information recognition model, determining first interactive information corresponding to the target live broadcast; wherein the first interactive information includes first interactive content and a first action identifier;

[0008] The first interactive information is sent to a target user terminal, so that the target user terminal performs a live interactive operation on a live broadcast interface based on the first interactive information in the target live broadcast.

[0009] According to another aspect of the present invention, a live interactive data processing device is provided, the device comprising:

[0010] A data acquisition module, used to acquire first explanation data generated in the target live broadcast;

[0011] An interactive information determination module, configured to determine first interactive information corresponding to the target live broadcast based on the first explanation data and a pre-trained information recognition model; wherein the first interactive information includes first interactive content and a first action identifier;

[0012] An information sending module is used to send the first interactive information to a target user terminal, so that the target user terminal performs a live interactive operation on a live broadcast interface based on the first interactive information in a target live broadcast. According to another aspect of the present invention, an electronic device is provided, comprising:

[0013] at least one processor; and

[0014] a memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the live interactive data processing method described in any embodiment of the present invention.

[0016] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the live interactive data processing method described in any embodiment of the present invention when executed.

[0017] According to another aspect of the present invention, a computer program product is provided, including a computer program, which, when executed by a processor, implements the live interactive data processing method described in any embodiment of the present invention.

[0018] The technical solution of the embodiment of the present invention obtains the first explanation data generated in the target live broadcast; based on the first explanation data and the pre-trained information recognition model, the first interactive information corresponding to the target live broadcast is determined; wherein the first interactive information includes the first interactive content and the first action identifier; since the interactive information is determined by the information recognition model, there is no need for the operator to determine the interactive content by watching the live broadcast, thereby improving the accuracy of the determined interactive content; and, by sending the first interactive information to the target user terminal, the target user terminal can perform live interactive operations in the live broadcast interface based on the first interactive information in the target live broadcast. The technical solution of this embodiment realizes interaction during the target live broadcast by sending the first interactive information to the target user terminal. Since the interactive process does not require manual operation, the accuracy and efficiency of the live broadcast interaction are improved, and the waste of human resources and time resources is reduced.

[0019] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 is a flow chart of a live interactive data processing method provided according to an embodiment of the present invention;

[0022] Figure 2 is a flowchart of another live interactive data processing method provided according to an embodiment of the present invention;

[0023] Figure 3 is a flowchart of another live interactive data processing method provided according to an embodiment of the present invention;

[0024] Figure 4 is a schematic diagram of an uplink process provided according to an embodiment of the present invention;

[0025] Figure 5 is a structural schematic diagram of a live interactive data processing device provided according to an embodiment of the present invention;

[0026] Figure 6 It is a structural diagram of an electronic device that implements the live interactive data processing method of an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "etc." and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0029] It should be noted that the collection, collection, updating, analysis, processing, use, transmission, storage and other aspects of user personal information involved in the technical solution of this disclosure are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information and network security.

[0030] Before introducing the technical solution, an example description of the application scenario can be given first. The technical solution can be applied to the scenario of performing live interactive operations in the target live broadcast. Exemplarily, the live interactive operation may include at least one of displaying interactive content, hiding interactive content, deleting interactive content, and pinning interactive content on the live broadcast interface. For example, after the anchor has explained the specific item, he or she usually verbally recites interactive instructions such as "up link", "down link", "display item", and "pin it", and these interactive instructions correspond to different live interactive operations. Through the technical solution, during the process of playing the target live broadcast, the interactive content and interactive actions matched by the target live broadcast can be determined through the pre-trained information recognition model, so that the target user end can perform corresponding interactive actions on the interactive content on the live broadcast interface to complete the live interactive operation. Based on the solution of this embodiment, there is no need for the participation of operating personnel, and the interactive content and interactive actions are quickly and accurately determined based on the information recognition model, thereby improving the accuracy and efficiency of live interaction and reducing the waste of human resources and time resources.

[0031] Figure 1 This is a flow chart of a live interactive data processing method provided according to an embodiment of the present invention. This embodiment is applicable to the case of live interactive on the live interface of the target user end. The method can be executed by a live interactive data processing device, which can be implemented in the form of hardware and / or software.

[0032] like Figure 1 As shown, the method of this embodiment may specifically include:

[0033] S110, obtaining first explanation data generated in the target live broadcast.

[0034] It should be noted that live broadcast is a form of real-time transmission of audio and video content through an Internet platform, and the target live broadcast is a live broadcast that needs to interact with the user end during the live broadcast. For example, the target live broadcast may include entertainment live broadcast, conference live broadcast, and e-commerce live broadcast.

[0035] In this embodiment, the first explanation data is the explanation data of the described object in the target live broadcast collected in the current collection period. The current collection period is set to the second level, for example, the explanation data is collected once per second. Optionally, the first explanation data can be at least one of voice data and text data.

[0036] In a specific implementation, the first explanation data generated in the target live broadcast can be obtained through the server. The server is connected to the target user terminal in communication, and the target user terminal is used to play the target live broadcast. The target user terminal includes at least one of a target live broadcast terminal and a target viewing terminal. The target live broadcast terminal is a terminal used to display to the anchor, and the target viewing terminal is a terminal used to display to the audience. For example, the target viewing terminal can be a mobile phone, computer, tablet, etc. of the audience watching the target live broadcast.

[0037] Optionally, obtaining the first explanation data generated in the target live broadcast includes: receiving live broadcast stream data corresponding to the target live broadcast sent by an information acquisition device; determining text data corresponding to the live broadcast stream data, and using the text data as the first explanation data; wherein the information acquisition device includes an audio acquisition device and at least one of a target live broadcast terminal communicatively connected to the audio acquisition device.

[0038] Specifically, the server can be connected to an audio acquisition device for acquiring audio information of the host. The server receives live streaming data sent by the audio acquisition device, and the live streaming data can be voice streaming data. In order to quickly identify the interactive information of the live interaction that needs to be executed, the first explanation data can be data in text form, so as to avoid the influence of factors such as the accent and tone of the host in the target live broadcast on the accuracy of information recognition. Therefore, after receiving the voice stream data, the server converts the voice stream data into text data, and uses the text data as the first explanation data.

[0039] Alternatively, the audio acquisition device is connected to the target live broadcast terminal, and the target live broadcast terminal obtains the voice stream data sent by the audio acquisition device. In order to reduce the workload of the server and complete the live interactive operation as soon as possible, the target live broadcast terminal converts the voice stream data into text data after receiving the voice stream data, and sends the text data as live stream data to the server, so that after the server receives the live stream data, it does not need to perform data conversion operations and uses the live stream data as the first explanation data.

[0040] Exemplarily, the target live broadcast end is communicatively connected with the anchor's microphone, the microphone collects voice stream data in real time, and periodically sends the collected voice stream data to the target anchor end at a period of one second. After the target live broadcast end receives the voice stream data corresponding to the current collection period, it can use a pre-built neural network model to convert the voice stream data into text data, and send the text data to the server.

[0041] This embodiment provides a method for obtaining first explanation data, and uses text data as the first explanation data, so as to avoid the influence of factors such as the accent and tone of the host in the target live broadcast on the accuracy of information recognition, which is conducive to accurately determining the interactive information.

[0042] S120. Determine first interactive information corresponding to the target live broadcast based on the first explanation data and a pre-trained information recognition model; wherein the first interactive information includes first interactive content and a first action identifier.

[0043] The information recognition model can be used to identify the attribute information of the description object corresponding to the first explanation data and the first action identifier; illustratively, the description object may include objects such as physical objects, texts, pictures, and animations. The first action identifier may include a display identifier, a delete identifier, and a top identifier, etc. The display identifier may include a pop-up window display identifier and a list display identifier. The attribute information is used to reflect the attributes of the description object. The attribute information includes at least one of the name, code, and owner information of the description object.

[0044] In a specific implementation, it can be determined whether the first explanation data is data in text format. If not, the first explanation data is converted into text format. In the case where the first explanation data is text data, the first explanation data can be input into a pre-trained information recognition model to obtain the output result of the information recognition model. The output result can reflect the attribute information and the first action identifier of the description object. Optionally, the content corresponding to the attribute information can be used as the first interactive content, so as to determine the first interactive information through the information recognition model.

[0045] S130. Send the first interactive information to the target user terminal, so that the target user terminal performs a live interactive operation on the live broadcast interface based on the first interactive information in the target live broadcast.

[0046] In this embodiment, the first interactive information may be sent to the target viewing terminal. Alternatively, in order to allow both the audience watching the live broadcast and the anchor recording the live broadcast to experience the live broadcast interaction, the first interactive information may be sent to the target viewing terminal and the target live broadcast terminal respectively.

[0047] In actual applications, the target user terminal performs a live interactive operation on the live interface based on the first interactive information by parsing the first interactive information, determining the first interactive content and the first action identifier, determining the interactive action that needs to be performed on the first interactive content through the first action identifier, and performing the interactive action on the first interactive content on the live interface to complete the live interactive operation.

[0048] For example, when the interactive action corresponding to the first action identifier is an upward link, a live broadcast card may be generated based on the first interactive content and displayed on the live broadcast interface. For example, the live broadcast card may be displayed in a pop-up window.

[0049] In order to simplify the operation and improve the convenience of live interaction, optionally, when there is no match between the first interaction information and the predetermined second interaction information, the first interaction information is sent to the target user terminal.

[0050] The second interactive information is interactive information determined based on the second explanation data generated by the target live broadcast and has been sent to the target user terminal. The second interactive information includes the second interactive content and the second action identifier.

[0051] It should be noted that, in the historical collection period, the second interactive information can be determined based on the second explanation data, and the second interactive information can be sent to the target user end to perform a live interactive operation. Among them, the second explanation data is the historical explanation data collected in the historical collection period. The second interactive information can be the same as the first interactive information, or it can be different from the first interactive information. In the case where the second interactive information is the same as the first interactive information, sending the first interactive information to the target user end will cause repeated sending and increase the workload. In order to avoid repeated sending, the first interactive information can be compared with the second interactive information before sending the first interactive information.

[0052] Specifically, the situation where the first interactive information does not match the predetermined second interactive information includes at least one of: the first interactive content is inconsistent with the second interactive content, and the first action identifier is inconsistent with the second action identifier. On the contrary, if the first interactive content is consistent with the second interactive content, and the first action identifier is consistent with the second action identifier, it means that the first interactive information matches the second interactive information. It should be noted that the second interactive information is the most recent interactive information sent to the target user terminal.

[0053] Furthermore, if the first interactive information does not match the second transmitted information, it means that the first interactive information and the second transmitted information are not repeated, and the first interactive information can be sent to the target user end to perform a live interactive operation based on the first interactive information. If the first interactive information matches the second transmitted information, it means that the target user end has received interactive information that is repeated with the first interactive information, and there is no need to send it again to avoid repeated sending.

[0054] This embodiment compares the first interactive information with the second interactive information before sending the first interactive information, so that the same interactive information is sent repeatedly, which simplifies the process of sending the interactive information and is conducive to improving the interactive efficiency.

[0055] The technical solution of the embodiment of the present invention obtains the first explanation data generated in the target live broadcast; based on the first explanation data and the pre-trained information recognition model, the first interactive information corresponding to the target live broadcast is determined; wherein the first interactive information includes the first interactive content and the first action identifier; since the interactive information is determined by the information recognition model, there is no need for the operator to determine the interactive content by watching the live broadcast, thereby improving the accuracy of the determined interactive content; and, by sending the first interactive information to the target user terminal, the target user terminal can perform live interactive operations in the live broadcast interface based on the first interactive information in the target live broadcast. The technical solution of this embodiment realizes interaction during the target live broadcast by sending the first interactive information to the target user terminal. Since the interactive process does not require manual operation, the accuracy and efficiency of the live broadcast interaction are improved, and the waste of human resources and time resources is reduced.

[0056] Figure 2 is a flowchart of another live interactive data processing method provided according to an embodiment of the present invention. After the first interactive information is sent to the target user terminal, this embodiment also includes: generating an action execution instruction when a preset indication condition is met, and sending the action execution instruction to the target viewing terminal, so that the target viewing terminal performs a live interactive operation on the live broadcast interface based on the first interactive information when receiving the action execution instruction. The explanations of the terms that are the same or corresponding to the above embodiments are not repeated here. Figure 2 As shown, the method includes:

[0057] S210: Acquire first explanation data generated in the target live broadcast.

[0058] S220. Determine first interactive information corresponding to the target live broadcast based on the first explanation data and a pre-trained information recognition model; wherein the first interactive information includes first interactive content and a first action identifier.

[0059] S230, sending the first interactive information to the target user terminal, generating an action execution instruction when the preset indication condition is met, and sending the action execution instruction to the target viewing terminal, so that the target viewing terminal, when receiving the action execution instruction, performs a live interactive operation on the live broadcast interface based on the first interactive information in the target live broadcast.

[0060] In the actual live broadcast process, the host usually explains the interactive content first, and then instructs the execution of the live broadcast interactive operation by verbally stating the interactive instructions after the comprehensive explanation. In order to enable the live broadcast interactive process to be executed at the host's pace, after sending the first interactive information to the target user end, it can be detected whether the preset instruction condition is met.

[0061] Optionally, the preset indication condition includes: receiving an interactive indication instruction sent by the target live broadcast end, and detecting that at least one of the preset action indicator words exists in the first explanation data.

[0062] Specifically, the target live broadcast terminal can obtain the voice stream data generated in the target live broadcast, and can promptly determine whether the anchor has issued an interactive instruction based on the received voice stream data. Therefore, the target live broadcast terminal can generate an interactive instruction when it is determined that the anchor has issued an interactive instruction, and send the interactive instruction to the server. When the server receives the interactive instruction, it generates an action execution instruction and sends the action execution instruction to the target viewing terminal, so that when the target viewing terminal receives the action execution instruction, it performs the live interactive operation based on the received first interactive information.

[0063] Among them, the target live broadcast end determines whether the anchor has issued an interactive instruction by parsing whether the voice stream data contains preset action indicators. Exemplarily, the preset action indicators include words such as link up, link down, and pin to the top. When the voice stream data contains at least one preset action indicator, it is determined that the anchor has issued an interactive instruction; when the voice stream data does not contain a preset action indicator, it is determined that the anchor has not issued an interactive instruction. Alternatively, the target live broadcast end determines whether the anchor has issued an interactive instruction by: when it is detected that the interactive execution control is triggered, it is determined that the anchor has issued an interactive instruction.

[0064] In order to avoid the failure to receive the interactive instruction due to the influence of the transmission signal, thereby affecting the timeliness of the live interaction, the server can detect whether the first explanation data contains preset action indicators after receiving the first explanation data. If the preset action indicators exist in the first explanation data, it can be determined that the host has issued an interactive instruction, and an action execution instruction can be directly generated and sent to the target viewing end, so that the target viewing end, when receiving the action execution instruction, performs the live interactive operation based on the received first interactive information in the target live broadcast.

[0065] Specifically, when the target viewing terminal receives the action execution instruction, it can determine that the first interactive information is the latest received interactive information; wherein the latest received interactive information includes the latest received interactive content and the latest received action identifier. If it is the latest received interactive information, the interactive action corresponding to the first action identifier can be executed on the first interactive content to perform the live interactive operation. If it is not the latest received interactive information, the interactive action corresponding to the latest received action identifier can be executed on the latest received interactive content.

[0066] This embodiment implements live interactive operations when the host issues interactive instructions by sending action execution instructions to the target user terminal, thereby ensuring that the interactive operations are synchronized with the host's oral instructions and improving the user experience.

[0067] In actual applications, during the live broadcast process, operators can watch the target live broadcast through the target live broadcast terminal. In order to improve the accuracy of the live broadcast interaction, the first interactive information is sent to the target user terminal, including: sending the first interactive information to the target live broadcast terminal, so that the target live broadcast terminal performs the interactive action corresponding to the first action identifier on the first interactive content according to the preset preview method; before generating the action execution instruction, it also includes: when receiving the adjustment instruction sent by the target live broadcast terminal, based on the adjustment instruction, updating the first interactive information.

[0068] In this embodiment, the target user terminal may include a target live broadcast terminal. Before receiving the first interactive information and not receiving the action execution instruction, the target live broadcast terminal may first preview the live interactive process corresponding to the first interactive information in a preset preview method, so that the operator can understand the effect of the live interactive operation. It should be noted that different interactive actions correspond to different preview methods. Exemplarily, the preview method corresponding to the upper link may be: using a pop-up window to display the live broadcast card corresponding to the first interactive content on the live broadcast interface according to a preset transparency value. The preview method corresponding to the lower link is: select the live broadcast card corresponding to the first interactive content displayed on the live broadcast interface through a dotted box. The corresponding preview method for pinning may be: pinning the live broadcast card corresponding to the first interactive content to the top of the live broadcast interface by flashing.

[0069] In a specific implementation, the operator can preview the effect of the live broadcast interaction to determine whether the first interactive information is the interactive effect that the anchor wants to perform. If so, no operation is required; if not, the target live broadcast end can be operated to generate an adjustment instruction and send the adjustment instruction to the server. The adjustment instruction may include the adjustment interaction information. The server can update the first interactive information to the adjustment interaction information based on the received adjustment instruction, thereby ensuring that the first interactive information can meet the anchor's interactive needs.

[0070] Before generating an action execution instruction, this embodiment previews the interaction process, so that the operating personnel can intuitively and conveniently understand the live interaction process in advance, and it is convenient to confirm whether the previewed live interaction process meets the interaction needs of the anchor; and if the interaction needs are not met, the first interaction information of the instruction is adjusted to improve the accuracy of the live interaction operation.

[0071] Figure 3 It is a flowchart of another live interactive data processing method provided according to an embodiment of the present invention. Based on the above embodiments, the specific implementation method of this embodiment to determine the first interactive information corresponding to the target live broadcast can be: based on the first explanation data and the information recognition model, determine the attribute information and the first action identifier of the descriptive object corresponding to the target live broadcast; determine the degree of match between the attribute information and at least one pre-stored alternative interactive content, and determine the alternative interactive content whose matching degree meets the preset matching conditions as the first interactive content. Among them, the explanations of the terms that are the same or corresponding to the above embodiments are not repeated here. Figure 3 As shown, the method includes:

[0072] S310: Acquire first explanation data generated in the target live broadcast.

[0073] S320: Determine the attribute information and the first action identifier of the description object corresponding to the target live broadcast based on the first explanation data and the information recognition model.

[0074] The attribute information includes at least one of the following information: the name, code and owner information of the object.

[0075] In this embodiment, before determining the attribute information of the descriptive object corresponding to the target live broadcast and the first action identifier, it can be determined whether the first explanation data is text data. If it is text data, the first explanation data can be used as input data of the information recognition model; if it is not text data, the first explanation data can be converted into text data, and the first explanation data can be updated based on the converted text data, and the updated first explanation data can be used as input data of the information recognition model.

[0076] Optionally, the information recognition model includes a text recognition model for information recognition based on preset prompt words; after ensuring that the first explanation data is text data, the implementation method of determining the attribute information and the first action identifier of the descriptive object corresponding to the target live broadcast based on the first explanation data and the information recognition model includes: determining the attribute information and the first action identifier of the descriptive object corresponding to the target live broadcast based on the first explanation data, a pre-trained text recognition model and preset prompt words corresponding to the text recognition model.

[0077] It should be noted that the text recognition model is a large model (Large Model) for identifying the attribute information and the first action identifier contained in the first explanation data. A large model, also known as a large language model or a basic model, refers to a machine learning model with a large parameter scale and a complex computing structure, such as the GPT model and the Gemini model.

[0078] In a specific implementation, preset prompt words corresponding to the text recognition model can be generated in advance. The preset prompt words can be words, phrases, questions or complete sentences. The preset prompt words are used to guide the text recognition model to identify the attribute information and the first action identifier reflected in the first explanation data. The preset prompt words include specific requirements, task instructions, examples and role settings. For example, the preset prompt words include "according to the detailed description below, the analysis contains the intention to top, or it says to top, and then analyzes the name of the item, the information of the owner of the item, and directly outputs the analysis results using the json structure; requirement 1 is that there is an item, and the corresponding output json structure example a; requirement 2 is that there is an item, but there is no item owner information, and the corresponding output json structure example b, requirement 3 is to directly output the json result, and other information is prohibited from being output; requirement 5 is that if there is an intention to top, or it says to top, add top to the original json structure example; otherwise, modify it to not top". Among them, json (JavaScript Object Notation, JS object notation) is a lightweight data exchange format.

[0079] In this embodiment, the first explanation data can be input into the text recognition model, and the text recognition model outputs the attribute information and the first action identifier of the description object corresponding to the target live broadcast based on the indication of the preset prompt word. This embodiment improves the accuracy and reliability of determining the attribute information and the first action identifier by using the preset prompt word and the text recognition model.

[0080] Alternatively, the information recognition model includes an attribute determination model and an intent classification model; after ensuring that the first explanation data is text data, based on the first explanation data and the information recognition model, the attribute information of the descriptive object corresponding to the target live broadcast and the first action identifier are determined, including: based on the first explanation data and the pre-trained attribute determination model, the attribute information of the descriptive object corresponding to the target live broadcast is determined; based on the first explanation data and the pre-trained intent classification model, the first action identifier is determined.

[0081] Among them, the attribute determination model includes a named entity recognition model, and the intent classification model includes a neural network model. The named entity recognition model is an important task in the field of natural language processing. Its main purpose is to identify and classify named entities in text, such as object names, place names, names of parties, time and date, etc. The intent classification model is used to identify the interactive intent contained in the first explanation data. Exemplarily, the intent classification model may include a Transformer model.

[0082] In a specific implementation, the first explanation data can be input into a pre-trained named entity recognition model, and the named entity recognition model determines the attribute information of the description object. The first explanation data can be input into an intention classification model, and the intention classification model outputs a first action identifier.

[0083] In this embodiment, the attribute information and the first action identifier are determined through the attribute determination model and the intention classification model, so that information recognition can be completed quickly. In addition, the training speed of the attribute determination model and the intention classification model is fast, which is conducive to improving the efficiency of information recognition.

[0084] In order to further improve the accuracy of the attribute information and the first action identifier, the attribute information and the first action identifier can be determined respectively according to the above two methods. Specifically, based on the preset prompt word, the first explanation data is input into the text recognition model to obtain the first output information and the first output identifier; the first explanation data is input into the attribute determination model and the intention classification model respectively, the second output information is input by the attribute determination model, and the second output identifier is output by the intention classification model. When the first output information is consistent with the second output information, and the first output identifier is consistent with the second output identifier, the first output information or the second output information is determined as the attribute information, and the first output identifier or the second output identifier is determined as the first action identifier. When the first output information is inconsistent with the second output information, or the first output identifier is inconsistent with the second output identifier, an early warning message can be generated and sent to the target live broadcast end, and the first interactive content and the first action identifier are determined based on the indication information fed back by the target live broadcast end. Alternatively, the first output information can be directly used as the attribute information, and the first output identifier can be used as the first action identifier. Thus, the attribute information and the first action identifier are determined through the output results of the two methods, and the accuracy and reliability of information recognition are further improved.

[0085] S330: Determine a matching degree between the attribute information and at least one pre-stored candidate interactive content, and determine the candidate interactive content whose matching degree satisfies a preset matching condition as the first interactive content.

[0086] The candidate interactive content may be the interactive content that needs to appear in the preset target live broadcast. For example, the candidate interactive content may include the display content of at least one item that needs to be linked in the target live broadcast.

[0087] In this embodiment, the degree of matching can be reflected by a numerical value. Exemplarily, the numerical values ​​corresponding to the matching degree include 0 and 1. It can be determined whether the candidate interactive content contains attribute information. If the attribute information is contained, the matching degree between the attribute information and the candidate interactive content is determined to be 1; otherwise, the matching degree between the attribute information and the candidate interactive content is determined to be 0. The preset matching condition can be that the numerical value corresponding to the matching degree is 1.

[0088] In addition, the degree of matching can also be reflected by the similarity between the attribute information and the candidate interactive content. The higher the similarity, the more matching the two are. Conversely, the lower the similarity, the less matching the two are. Optionally, determining the degree of matching between the attribute information and at least one pre-stored candidate interactive content includes: generating a description vector for the description object based on the attribute information and the first explanation data, and determining the candidate content vector corresponding to each candidate interactive content; for each candidate interactive content, determining the vector similarity between the description vector and the candidate content vector corresponding to the candidate interactive content, and determining the vector similarity as the degree of matching between the candidate interactive content and the attribute information.

[0089] In this embodiment, the attribute information and the first explanation data can be spliced ​​to obtain comprehensive description information of the description object, and the description information can be converted into an embedded vector through natural language processing technology, and the embedded vector is used as the description vector. The description vector can reflect the characteristics and attributes of the description object. In addition, for each candidate interactive content, natural language processing technology can be used to generate a corresponding candidate content vector. The vector similarity between the description vector and each candidate content vector is determined by a preset similarity determination method. Exemplarily, the preset similarity determination method may include a cosine similarity determination method, a Euclidean distance determination method, and a Manhattan distance determination method. The vector similarity is determined as the degree of matching between the candidate interactive content and the attribute information.

[0090] In this embodiment, the degree of match between the candidate interactive content and the attribute information is determined by vector similarity, thereby improving the accuracy of the determined degree of match; and, based on a description vector that can comprehensively reflect the description object, the vector similarity is determined, thereby avoiding the situation where the degree of match is determined in a biased manner due to incomplete information.

[0091] S340: Send the first interactive content and the first action identifier to the target user terminal, so that the target user terminal performs a live interactive operation on the live broadcast interface based on the first interactive content and the first action identifier in the target live broadcast.

[0092] In this embodiment, the first interactive content and the first action identifier are sent to the target user terminal using WebSocket technology. WebSocket is a protocol for full-duplex communication on a single transmission control protocol connection, which can make data exchange between the target user terminal and the server terminal simpler, allowing the server terminal to actively send data to the target user terminal and perform two-way data transmission.

[0093] This embodiment uses WebSocket technology to push content to the target user end to perform live interactive operations, ensuring that the first interactive content and the first action identifier can be displayed to the target user end in real time and accurately, thereby improving the user's interactive experience.

[0094] The embodiments corresponding to the live interactive data processing method are described in detail above. In order to make the technical scheme of the method further clear to those skilled in the art, the following is a specific description taking the link operation on the recommended items as an example.

[0095] Figure 4 is a schematic diagram of an uplink process provided according to an embodiment of the present invention; Figure 4As shown, during the target live broadcast process, the video acquisition device collects the video data corresponding to the target live broadcast in real time, and translates the video data into language stream data through the simultaneous interpretation technology; or, the language stream data in the target live broadcast is collected in real time by the language acquisition device. The language stream data contains at least one of the name, code, ownership information and specifications of the recommended item. The language stream data is converted into text data as the first explanation data corresponding to the current acquisition cycle. By inputting the first explanation data into a pre-trained text recognition model, the first explanation data is semantically recognized to obtain the attribute information corresponding to the recommended item, and the attribute information is matched with the pre-stored alternative interactive content, so as to determine the alternative interactive content matching the attribute information as the first interactive content. Exemplarily, the first interactive content includes the item identification, physical picture, value attribute value, material, ownership information and size of the recommended item. In addition, the first action identification of the recommended action of the recommended item is determined through the output information of the text recognition model; for example, the recommended action may include a top recommendation or a pop-up window recommendation. The first interactive content and the first action identifier are sent to the target user terminal using WebSocket technology; wherein the target user terminal includes a target live broadcast terminal and a target viewing terminal. Furthermore, when it is detected that the first explanation data includes an "up link" instruction, an action execution instruction is generated; for example, the action execution instruction may be an instruction to pop up an item or an instruction to pin an item. Furthermore, the action execution instruction is sent to the target user terminal. After receiving the action execution instruction, the target user terminal generates a live broadcast card based on the first interactive content, and displays the generated live broadcast card according to the recommended action corresponding to the first action identifier, so as to complete the interactive operation of the up link of the first interactive content.

[0096] This embodiment determines the first interactive content and the first action identifier through a text recognition model, thereby achieving fast and accurate object recognition, instruction extraction and semantic analysis; and, through language recognition technology, converts voice stream data into text, which can overcome the dialect and accent recognition problems in voice recognition; finally, through real-time interaction technology, it is possible to automatically make real-time recommendations for recommended items given by the host in the live broadcast room scenario, thereby reducing labor costs, which is conducive to improving live broadcast efficiency and reducing live broadcast operating costs.

[0097] Figure 5 1 is a schematic diagram of the structure of a live interactive data processing device provided according to an embodiment of the present invention, and the device is used to execute the live interactive data processing method provided by any of the above embodiments. The device and the live interactive data processing method of the above embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the live interactive data processing device, please refer to the embodiments of the above live interactive data processing method. Figure 5 As shown, the device comprises:

[0098] The data acquisition module 10 is used to acquire the first explanation data generated in the target live broadcast;

[0099] The interactive information determination module 11 is used to determine the first interactive information corresponding to the target live broadcast based on the first explanation data and the pre-trained information recognition model; wherein the first interactive information includes the first interactive content and the first action identifier;

[0100] The information sending module 12 is used to send the first interactive information to the target user end, so that the target user end performs a live interactive operation on the live interface based on the first interactive information in the target live broadcast. Based on any optional technical solution in the embodiment of the present invention, the information sending module 12 optionally includes:

[0101] A first information sending submodule, configured to send the first interaction information to a target user terminal when there is no match between the first interaction information and the predetermined second interaction information;

[0102] Among them, the second interactive information is interactive information determined based on the second explanation data generated by the target live broadcast and has been sent to the target user end, and the second interactive information includes second interactive content and second action identifier; the situation where the first interactive information does not match the predetermined second interactive information includes: at least one of the first interactive content and the second interactive content are inconsistent, and the first action identifier and the second action identifier are inconsistent.

[0103] On the basis of any optional technical solution in the embodiment of the present invention, optionally, the target user terminal includes at least one of a target live broadcast terminal and a target viewing terminal, and the device further includes:

[0104] An instruction generation module is used to generate an action execution instruction after sending the first interactive information to the target user terminal and, if a preset indication condition is met, send the action execution instruction to the target viewing terminal, so that the target viewing terminal performs a live interactive operation on the live broadcast interface based on the first interactive information in the target live broadcast when receiving the action execution instruction;

[0105] The preset indication conditions include: receiving an interactive indication instruction sent by the target live broadcast terminal, and detecting that at least one of the preset action indication words exists in the first explanation data.

[0106] Based on any optional technical solution in the embodiment of the present invention, optionally, the information sending module 12 includes:

[0107] The second information sending submodule is used to send the first interactive information to the target live broadcast end, so that the target live broadcast end performs the interactive action corresponding to the first action identifier on the first interactive content according to the preset preview mode;

[0108] Instruction generation module, including:

[0109] The information updating submodule is used to update the first interactive information based on the adjustment instruction when an adjustment instruction sent by the target live broadcast end is received before generating the action execution instruction.

[0110] Based on any optional technical solution in the embodiment of the present invention, optionally, the interactive information determination module 11 includes:

[0111] The interactive information determination submodule is used to determine the attribute information and the first action identifier of the description object corresponding to the target live broadcast based on the first explanation data and the information recognition model; wherein the attribute information includes at least one of the name, code and ownership information of the description object;

[0112] The matching degree determination submodule is used to determine the matching degree between the attribute information and at least one pre-stored candidate interactive content, and determine the candidate interactive content whose matching degree meets the preset matching condition as the first interactive content.

[0113] On the basis of any optional technical solution in the embodiment of the present invention, optionally, the information recognition model includes a text recognition model for information recognition based on preset prompt words;

[0114] The interactive information determination submodule includes:

[0115] The first determination unit is used to determine the attribute information and the first action identifier of the description object corresponding to the target live broadcast based on the first explanation data, the pre-trained text recognition model and the preset prompt words corresponding to the text recognition model.

[0116] On the basis of any optional technical solution in the embodiment of the present invention, optionally, the information recognition model includes an attribute determination model and an intent classification model;

[0117] The interactive information determination submodule includes:

[0118] A second determination unit is used to determine attribute information of a description object corresponding to a target live broadcast based on the first explanation data and a pre-trained attribute determination model;

[0119] An identifier determining unit, configured to determine a first action identifier based on the first explanation data and a pre-trained intention classification model;

[0120] Among them, the attribute determination model includes a named entity recognition model, and the intent classification model includes a neural network model.

[0121] Based on any optional technical solution in the embodiment of the present invention, optionally, the matching degree determination submodule includes:

[0122] A vector generating unit, configured to generate a description vector for the description object based on the attribute information and the first explanation data, and determine a candidate content vector corresponding to each candidate interactive content;

[0123] The similarity determination unit is used to determine the vector similarity between the description vector and the candidate content vector corresponding to the candidate interactive content for each candidate interactive content, and determine the vector similarity as the matching degree between the candidate interactive content and the attribute information.

[0124] Based on any optional technical solution in the embodiment of the present invention, optionally, the data acquisition module 10 includes:

[0125] The data receiving submodule is used to receive the live streaming data corresponding to the target live streaming sent by the information collection device;

[0126] The data determination submodule is used to determine the text data corresponding to the live streaming data and use the text data as the first explanation data; wherein the information collection device includes at least one of an audio collection device and a target live broadcast terminal that is communicatively connected to the audio collection device.

[0127] The technical solution of the embodiment of the present invention obtains the first explanation data generated in the target live broadcast; based on the first explanation data and the pre-trained information recognition model, the first interactive information corresponding to the target live broadcast is determined; wherein the first interactive information includes the first interactive content and the first action identifier; since the interactive information is determined by the information recognition model, there is no need for the operator to determine the interactive content by watching the live broadcast, thereby improving the accuracy of the determined interactive content; and, by sending the first interactive information to the target user terminal, the target user terminal can perform live interactive operations in the live broadcast interface based on the first interactive information in the target live broadcast. The technical solution of this embodiment realizes interaction during the target live broadcast by sending the first interactive information to the target user terminal. Since the interactive process does not require manual operation, the accuracy and efficiency of the live broadcast interaction are improved, and the waste of human resources and time resources is reduced.

[0128] It is worth noting that in the embodiment of the above-mentioned live interactive data processing device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0129] Figure 6Schematic diagram of the structure of an electronic device that implements the live interactive data processing method of an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0130] like Figure 6 As shown, the electronic device 20 includes at least one processor 21, and a memory connected to the at least one processor 21, such as a read-only memory (ROM) 22, a random access memory (RAM) 23, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 21 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 22 or the computer program loaded from the storage unit 28 to the random access memory (RAM) 23. In RAM23, various programs and data required for the operation of the electronic device 20 can also be stored. The processor 21, ROM22 and RAM23 are connected to each other through a bus 24. An input / output (I / O) interface 25 is also connected to the bus 24.

[0131] A number of components in the electronic device 20 are connected to the I / O interface 25, including: an input unit 26, such as a keyboard, a mouse, etc.; an output unit 27, such as various types of displays, speakers, etc.; a storage unit 28, such as a disk, an optical disk, etc.; and a communication unit 29, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 29 allows the electronic device 20 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0132] The processor 21 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 21 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 21 executes the various methods and processes described above, such as the live interactive data processing method.

[0133] In some embodiments, the live interactive data processing method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 28. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 20 via the ROM 22 and / or the communication unit 29. When the computer program is loaded into the RAM 23 and executed by the processor 21, one or more steps of the live interactive data processing method described above may be performed. Alternatively, in other embodiments, the processor 21 may be configured to execute the live interactive data processing method in any other appropriate manner (e.g., by means of firmware).

[0134] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0135] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0136] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0137] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0138] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0139] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0140] This embodiment also provides a computer program product, including a computer program, which, when executed by a processor, implements the live interactive data processing method provided in any embodiment of the present application.

[0141] The computer program product may be implemented in a computer program code for performing the operation of the present invention written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​and conventional procedural programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0142] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0143] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A live interactive data processing method, characterized in that: include: Obtain the first explanation data generated in the target live broadcast; Based on the first explanation data and the pre-trained information recognition model, determining first interactive information corresponding to the target live broadcast; wherein the first interactive information includes first interactive content and a first action identifier; The first interactive information is sent to a target user terminal, so that the target user terminal performs a live interactive operation on a live broadcast interface based on the first interactive information during the target live broadcast.

2. The method according to claim 1, characterized in that The sending the first interactive information to the target user terminal includes: In the case where the first interaction information does not match the predetermined second interaction information, sending the first interaction information to the target user terminal; Among them, the second interactive information is interactive information determined based on the second explanation data generated by the target live broadcast and has been sent to the target user terminal, and the second interactive information includes second interactive content and second action identifier; the situation where the first interactive information does not match the predetermined second interactive information includes: the first interactive content is inconsistent with the second interactive content, and the first action identifier is inconsistent with the second action identifier.

3. The method according to claim 1, characterized in that The target user terminal includes at least one of a target live broadcast terminal and a target viewing terminal. After sending the first interactive information to the target user terminal, the method further includes: When the preset indication condition is met, an action execution instruction is generated, and the action execution instruction is sent to the target viewing terminal, so that the target viewing terminal performs a live broadcast interactive operation on the live broadcast interface based on the first interactive information in the target live broadcast when receiving the action execution instruction; Among them, the preset indication condition includes: receiving an interactive indication instruction sent by the target live broadcast end, and detecting that at least one of the preset action indicator words exists in the first explanation data.

4. The method according to claim 3, characterized in that The sending the first interactive information to the target user terminal includes: Sending the first interactive information to the target live broadcast end, so that the target live broadcast end performs the interactive action corresponding to the first action identifier on the first interactive content according to a preset preview mode; Before generating the action execution instruction, the method further includes: When an adjustment instruction sent by the target live broadcast end is received, the first interactive information is updated based on the adjustment instruction.

5. The method according to claim 1, characterized in that The determining, based on the first explanation data and the pre-trained information recognition model, first interactive information corresponding to the target live broadcast includes: Based on the first explanation data and the information recognition model, determine the attribute information and the first action identifier of the description object corresponding to the target live broadcast; wherein the attribute information includes at least one of the name, code and attribution information of the description object; The matching degree between the attribute information and at least one pre-stored candidate interactive content is determined, and the candidate interactive content whose matching degree satisfies a preset matching condition is determined as the first interactive content.

6. The method according to claim 5, characterized in that The information recognition model includes a text recognition model for information recognition based on preset prompt words; The determining, based on the first explanation data and the information recognition model, the attribute information and the first action identifier of the description object corresponding to the target live broadcast includes: Based on the first explanation data, the pre-trained text recognition model and the preset prompt words corresponding to the text recognition model, the attribute information of the description object corresponding to the target live broadcast and the first action identifier are determined.

7. The method according to claim 5, characterized in that The information recognition model includes an attribute determination model and an intention classification model; The determining, based on the first explanation data and the information recognition model, the attribute information and the first action identifier of the description object corresponding to the target live broadcast includes: Determine attribute information of a description object corresponding to the target live broadcast based on the first explanation data and the pre-trained attribute determination model; Determining the first action identifier based on the first explanation data and the pre-trained intent classification model; Among them, the attribute determination model includes a named entity recognition model, and the intent classification model includes a neural network model.

8. The method according to claim 5, characterized in that Determining the degree of matching between the attribute information and at least one pre-stored candidate interactive content includes: Based on the attribute information and the first explanation data, a description vector for the description object is generated, and a candidate content vector corresponding to each candidate interactive content is determined; For each of the candidate interactive contents, a vector similarity between the description vector and a candidate content vector corresponding to the candidate interactive content is determined, and the vector similarity is determined as a matching degree between the candidate interactive content and the attribute information.

9. The method according to claim 1, characterized in that: The obtaining of the first explanation data generated in the target live broadcast includes: Receiving live streaming data corresponding to the target live broadcast sent by an information collection device; Determine text data corresponding to the live streaming data, and use the text data as the first explanation data; wherein the information acquisition device includes at least one of an audio acquisition device and a target live broadcast terminal that is communicatively connected to the audio acquisition device.

10. A live interactive data processing device, characterized in that: include: A data acquisition module, used to acquire first explanation data generated in the target live broadcast; An interactive information determination module, configured to determine first interactive information corresponding to the target live broadcast based on the first explanation data and a pre-trained information recognition model; wherein the first interactive information includes first interactive content and a first action identifier; An information sending module is used to send the first interactive information to a target user terminal, so that the target user terminal performs a live interactive operation on a live broadcast interface based on the first interactive information in a target live broadcast.

11. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the live interactive data processing method described in any one of claims 1-9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the live interactive data processing method described in any one of claims 1-9 when executed.

13. A computer program product, comprising a computer program, which, when executed by a processor, implements the live interactive data processing method according to any one of claims 1-9.

Images