Digital anchor interaction method and device
By querying and updating the historical interactive data of digital anchors and using character setting data, the problem of difficult to remember long conversation content of large language models is solved, improving the accuracy and consistency of virtual anchor responses, and improving user experience.
Patent Information
- Application Number
- CN202410933109.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-11
- Publication Date
- 2024-11-15
AI Technical Summary
The large language model finds it difficult to remember the long conversation content when generating virtual anchor replies, resulting in inconsistent dialogue or lack of logic, and it is difficult to capture long-term dependencies, affecting the user experience.
By querying the historical interactive data between the digital anchor and the target user, using a large language model to generate reply information based on the interaction information and memory data, update the memory data to improve the targetedness and accuracy of the reply, including the processing of inductive memory data and temporary memory data. And use character setting data to generate replies when memory data is lacking.
It improves the accuracy and consistency of digital anchor responses, reduces inconsistent dialogue content, and enhances user experience.
Smart Images

Figure CN118964544A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the field of data processing technology, and in particular, to a digital anchor interaction method, device, computer equipment, computer-readable storage medium, and computer program product. Background Art
[0002] With the development of generative large language models, current virtual anchors can now complete the live broadcast process without the cooperation of real people, but through large language models to achieve the entire live broadcast process. These virtual anchors that mainly rely on artificial intelligence technology are also called digital anchors.
[0003] However, since the current prediction results of the large language model only depend on the current input, when the conversation context is long, the large language model cannot fully remember the previous conversation content; moreover, as the conversation progresses, previous information may be overwritten by new information, making it difficult for the large language model to accurately remember the entire conversation history; in addition, the large language model usually processes long sequence data by limiting the length of the context or using a truncation mechanism, making it difficult for the large language model to capture long-term dependencies and have difficulty understanding long conversation contexts. Due to the above problems of the large language model, when the large language model generates the digital anchor's response, the conversation content is often inconsistent or lacks logic, and the digital anchor may even not remember the audience with whom he has interacted, resulting in a reduced audience experience.
[0004] It should be noted that the above content is not necessarily prior art, nor is it intended to limit the scope of patent protection of this application. Summary of the invention
[0005] The embodiments of the present application provide a digital anchor interaction method, apparatus, computer equipment, computer-readable storage medium, and computer program product to solve or alleviate one or more of the technical problems raised above.
[0006] One aspect of an embodiment of the present application provides a digital anchor interaction method, the method comprising:
[0007] When receiving the interaction information of the target user, query whether there is memory data of the target user by the digital anchor, wherein the memory data is obtained according to the historical interaction data between the target user and the digital anchor;
[0008] In the case where the memory data exists, the large language model corresponding to the digital anchor is used to generate interactive reply information of the target user according to the interactive information and the memory data.
[0009] Optionally, the memory data includes inductive memory data and temporary memory data, the inductive memory data is generated according to a preset amount of historical interaction data, and the temporary memory data is the remaining historical interaction data except the inductive memory data.
[0010] Optionally, the method further comprises:
[0011] The interactive information and the interactive reply information constitute a complete conversation and are included in the temporary memory data to obtain updated temporary memory data;
[0012] Determining whether the number of the updated temporary memory data reaches the preset number;
[0013] When the number of the updated temporary memory data reaches the preset number, generating new inductive memory data according to the updated temporary memory data and the inductive memory data;
[0014] The memory data is updated according to the new summarized memory data.
[0015] Optionally, the method further comprises:
[0016] When the number of the updated temporary memory data is less than the preset number, the memory data is updated according to the updated temporary memory data and the summarized memory data.
[0017] Optionally, the method further comprises:
[0018] If the memory data cannot be found, obtaining the character setting data of the digital anchor;
[0019] The interactive response information is generated by using the large language model according to the interactive information and the character setting data.
[0020] Optionally, the using of the large language model corresponding to the digital anchor to generate the interactive reply information of the target user according to the interactive information and the memory data includes:
[0021] Obtaining character setting data of the digital anchor;
[0022] The large language model is used to generate interactive response information of the target user according to the interactive information, the memory data and the character setting data.
[0023] Optionally, the step of using the large language model corresponding to the digital anchor to generate the interactive reply information of the target user according to the interactive information and the memory data includes:
[0024] generating a prompt based on the memory data;
[0025] The prompt and the interaction information are input into the large language model, so as to utilize the large language model to generate the interaction response information according to the prompt and the interaction information.
[0026] Optionally, the method further comprises:
[0027] Converting the interactive response information into voice information;
[0028] Obtaining lip shape data corresponding to the digital anchor;
[0029] The voice information is outputted by the digital anchor according to the lip shape data.
[0030] Another aspect of the embodiments of the present application provides a digital anchor interaction device, the device comprising:
[0031] A query module, configured to query whether there is memory data of the target user by the digital anchor when receiving the interaction information of the target user, wherein the memory data is obtained according to the historical interaction data between the target user and the digital anchor;
[0032] A reply module is used to generate interactive reply information of the target user according to the interactive information and the memory data by using the large language model corresponding to the digital anchor when the memory data exists.
[0033] Another aspect of an embodiment of the present application provides a computer device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein: the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method described above.
[0034] Another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the method described above is implemented.
[0035] Another aspect of an embodiment of the present application provides a computer program product, including a computer program, which implements the method described above when executed by a processor.
[0036] The above technical solution adopted in the embodiment of the present application may have the following advantages:
[0037] By obtaining the digital anchor's memory data of the target user based on the historical interaction data between the target user and the digital anchor, upon receiving the interaction information of the target user, the digital anchor's memory data of the target user is queried, and the large language model corresponding to the digital anchor is used to generate the target user's interaction reply information based on the interaction information and the memory data. This allows the digital anchor to generate reply information based on the historical interaction data with the user, thereby enabling the digital anchor to "remember" the audience with whom the interaction has occurred, thereby improving the pertinence and accuracy of reply generation, reducing the inconsistency or lack of logic in conversation content, and improving user experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] The accompanying drawings exemplarily illustrate the embodiments and constitute a part of the specification, and together with the text description of the specification, are used to explain the exemplary implementation of the embodiments. The embodiments shown are for illustrative purposes only and do not limit the scope of the claims. In all drawings, the same reference numerals refer to similar but not necessarily identical elements.
[0039] Figure 1 A diagram schematically shows an operating environment of a digital anchor interaction method according to Embodiment 1 of the present application;
[0040] Figure 2 A flowchart of a digital anchor interaction method according to Embodiment 1 of the present application is schematically shown;
[0041] Figure 3 Schematically shows Figure 2 Flow chart of sub-steps of step S202;
[0042] Figure 4 A flowchart of newly added steps of the digital anchor interaction method according to the first embodiment of the present application is schematically shown;
[0043] Figure 5 Schematically shows Figure 2 Another sub-step flow chart of step S202;
[0044] Figure 6 A flowchart of another newly added step of the digital anchor interaction method according to the first embodiment of the present application is schematically shown;
[0045] Figure 7 A flowchart of another newly added step of the digital anchor interaction method according to the first embodiment of the present application is schematically shown;
[0046] Figure 8 A diagram schematically shows an application example of the digital anchor interaction method according to the first embodiment of the present application;
[0047] Fig. 9 A block diagram of a digital anchor interaction device according to Embodiment 2 of the present application is schematically shown; and
[0048] Fig.10 The hardware architecture diagram of the computer device according to the third embodiment of the present application is schematically shown. DETAILED DESCRIPTION
[0049] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.
[0050] It should be noted that the descriptions involving "first", "second", etc. in the embodiments of the present application are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.
[0051] In the description of the present application, it should be understood that the numerical labels before the steps do not indicate the order in which the steps are executed, but are only used to facilitate the description of the present application and to distinguish each step, and therefore should not be understood as a limitation on the present application.
[0052] First, the following terms are explained:
[0053] A virtual anchor refers to an anchor who uses a virtual image to post content on video websites. It is a form of anchor that uses a virtual image and voice to publish content on video websites. It combines computer technology, network technology and artificial intelligence technology to bring a new audio-visual experience to the audience.
[0054] A digital anchor, also known as a digital human anchor, is a virtual image generated based on artificial intelligence technology. It is a form of anchor used for content dissemination and interaction on media platforms such as live broadcasts. It is a type of virtual anchor.
[0055] The Big Language Model is a deep learning model trained on massive text data, with a large parameter scale and powerful natural language processing capabilities. It can deeply understand the meaning of text, generate natural language text, and perform various natural language tasks. It is one of the important technologies in the field of artificial intelligence.
[0056] Secondly, in order to facilitate those skilled in the art to understand the technical solutions provided in the embodiments of the present application, the relevant technologies are described below:
[0057] In the related art, digital anchors implemented through large language models have difficulty in understanding the context of long conversations, as large language models find it difficult to capture long-term dependencies. Therefore, when generating the digital anchor's responses, the conversation content is often inconsistent or lacks logic. The digital anchor may also not remember the audience with whom he has interacted, resulting in a decline in the audience experience.
[0058] To this end, the embodiment of the present application provides a digital anchor interaction technical solution. In this technical solution, the digital anchor's reply can be generated based on the user's historical interaction data, reducing the inconsistency or lack of logic in the conversation content, improving the pertinence and accuracy of the reply generation, and thus improving the user experience. See below for details.
[0059] Finally, for ease of understanding, an exemplary operating environment is provided below.
[0060] like Figure 1 As shown, the operating environment diagram includes a service platform 2, anchor terminals (4A, 4B, ..., 4M), and audience terminals (6A, 6B, ..., 6N). In the live broadcast scenario, the anchor terminal (4A, 4B, ..., 4M) logs in to the service platform 2, and the service platform 2 pushes the live broadcast data to the audience terminal (6A, 6B, ..., 6N) in real time.
[0061] Service platform 2 can provide live broadcast service, which can be a single server, a server cluster or a cloud computing service center.
[0062] The anchor terminal (4A, 4B, ..., 4M) is used to generate live broadcast data in real time and perform streaming operations on the live broadcast data. The live broadcast data may include audio data or video data. The anchor terminal may be an electronic device such as a smart phone or a tablet computer. Of course, the anchor terminal may be a virtual computing instance in the service platform 2.
[0063] The audience terminal (6A, 6B, ..., 6N) can be configured to receive the live broadcast data of the anchor terminal in real time. The audience terminal (6A, 6B, ..., 6N) can be any type of computing device, such as a smart phone, a tablet device, a laptop computer, a smart TV, a car terminal, etc. The audience terminal (6A, 6B, ..., 6N) can have a built-in browser or a special program, and receive the live broadcast data through the browser or the special program to output content to the user. The content may include video, audio, comments, text data and / or the like.
[0064] The audience terminal (6A, 6B, ..., 6N) may include a player. The player outputs (e.g., displays, presents) content to the user. The content may include video, audio, comments, text data and / or the like. The audience terminal (6A, 6B, ..., 6N) may include an interface, which may include an input element (touch screen). For example, the input element may be configured to receive user instructions, which may enable the audience terminal (6A, 6B, ..., 6N) to perform various operations, such as sending a bullet screen, entering a comment, giving a gift, etc.
[0065] The anchor terminals (4A, 4B, ..., 4M), the audience terminals (6A, 6B, ..., 6N) and the service platform 2 may be connected via a network. The network may include various network devices, such as routers, switches, multiplexers, hubs, modems, bridges, repeaters, firewalls and / or proxy devices, etc. The network may include physical links, such as coaxial cable links, twisted pair cable links, optical fiber links, combinations thereof and / or the like. The network may include wireless links, such as cellular links, satellite links, Wi-Fi links and / or the like.
[0066] It should be noted that the number of the host terminals and the audience terminals in the figure is only for illustration and is not used to limit the scope of patent protection of this application. According to actual conditions, there can be any number of host terminals and audience terminals.
[0067] The following uses the service platform as the execution subject to introduce the technical solution of the present application through multiple embodiments. It should be noted that these embodiments can be implemented in a variety of different forms and should not be interpreted as being limited to the embodiments described here.
[0068] Embodiment 1
[0069] Figure 2 The flowchart of the digital anchor interaction method according to the first embodiment of the present application is schematically shown.
[0070] like Figure 2 As shown, the digital anchor interaction method may include steps S200 to S202, wherein:
[0071] Step S200, when receiving the interaction information of the target user, query whether there is any memory data of the digital anchor on the target user, and the memory data is obtained according to the historical interaction data between the target user and the digital anchor.
[0072] Step S202, when there is memory data, using the large language model corresponding to the digital anchor to generate interactive reply information of the target user according to the interactive information and the memory data.
[0073] The digital anchor interaction method provided in the present embodiment obtains the digital anchor's memory data of the target user based on the historical interaction data between the target user and the digital anchor, queries the digital anchor's memory data of the target user when receiving the interaction information of the target user, and uses the large language model corresponding to the digital anchor to generate the target user's interaction reply information based on the interaction information and the memory data. The digital anchor can generate reply information based on the historical interaction data with the user, so that the digital anchor can "remember" the audience with whom the interaction has occurred, thereby improving the pertinence and accuracy of reply generation, reducing the inconsistency or lack of logic in the conversation content, and improving the user experience.
[0074] The following combination Figure 2 , each step in steps S200 to S202 and other optional steps are described in detail.
[0075] Step S200 ,When receiving the interactive information of the target user, it is queried whether there is memory data of the target user by the digital anchor, and the memory data is obtained according to the historical interactive data between the target user and the digital anchor.
[0076] The target user's interactive information may include, but is not limited to, text messages (such as sending bullet comments), gift messages (such as sending gifts to digital anchors), system messages (such as liking, collecting, forwarding), or voice messages. Similarly, the historical interactive data between the target user and the digital anchor may include, but is not limited to, text interaction data, gift interaction data, system interaction data, or voice interaction data.
[0077] The memory data can be directly the historical interaction data between the target user and the digital anchor. For example, if the target user and the digital anchor have two historical interaction data, the memory data can be the two historical interaction data. Optionally, the memory data can also be obtained by processing the historical interaction data. For example, if the historical interaction data is text information, the memory data can be the data obtained after removing stop words from the text information; for another example, if the historical interaction data is gift interaction data, the memory data can be the gift interaction data converted into data related to the digital anchor's favorability or intimacy with the target user.
[0078] In an optional embodiment, the memory data includes inductive memory data and temporary memory data, wherein the inductive memory data is generated based on a preset amount of historical interaction data, and the temporary memory data is the remaining historical interaction data except the inductive memory data.
[0079] That is, the memory data can be divided into two parts, one part is the data obtained by inductive processing, and the other part is the data without inductive processing.
[0080] The preset number can be configured according to actual needs, for example, it can be 20. When the number of historical interaction data reaches 20, the 20 historical interaction data are summarized and processed to obtain one piece of summarized memory data.
[0081] When generating inductive memory data based on a preset amount of historical interaction data, the text interaction data and voice interaction data in the historical interaction data may be summarized through natural language processing methods to obtain summarized information; the gift interaction data and system interaction data in the historical interaction data may be determined according to certain calculation rules to determine information on the digital anchor's favorability toward the target user; finally, the summarized information and information on favorability may be combined to obtain the inductive memory data.
[0082] For example, for the inductive processing of text interaction data, natural language processing methods can be used to perform part-of-speech analysis, syntactic analysis, semantic analysis and other analysis and processing to extract the main key information. For another example, for the inductive processing of gift interaction data, several levels of favorability between digital anchors and users can be set in advance (such as audience-ordinary friends-close friends), each level corresponds to a different score, and different gifts are set in advance to correspond to different favorability scores. The target user's gift interaction data and the corresponding scores are scored, and the information on the digital anchor's favorability level for the user is determined based on the cumulative scoring results.
[0083] When summarizing and processing text interaction data and voice interaction data, it is also possible to conduct an emotional analysis, and obtain information about the digital anchor's favorability toward the user based on the results of the analysis. In addition, when summarizing and processing historical interaction data to obtain summarized memory data, it is also possible to summarize in combination with the character setting of the digital anchor itself. For example, for a digital anchor whose character setting is food, the historical interaction data can be summarized and processed in combination with food, that is, more attention is paid to information about "food". For another example, for a digital anchor whose character setting is games, the historical interaction data can be summarized and processed in combination with games, that is, more attention is paid to information about "games", and so on.
[0084] In the above optional embodiment, the memory data includes inductive memory data and temporary memory data. The inductive memory data is generated based on a preset amount of historical interaction data. The historical interaction data can be at least partially summarized to obtain more concise and refined memory data, thereby improving the efficiency of the large language model corresponding to the digital anchor in generating replies based on the memory data.
[0085] Step S202 ,In the presence of memory data, the large language model corresponding to the digital anchor is used to generate the target user’s interactive response information based on the interactive information and memory data.
[0086] In an optional embodiment, if Figure 3 As shown, step S202 may further include steps S600 to S602:
[0087] Step S600, generating a prompt according to the memory data.
[0088] Step S602, inputting the prompt and interaction information into the large language model, so as to utilize the large language model to generate interaction reply information according to the prompt and interaction information.
[0089] Specifically, the service platform generates prompts based on the memory data, and inputs the interactive information and prompts into the large language model corresponding to the digital anchor, so that the large language model generates interactive reply information for the target user based on the interactive information under the prompts.
[0090] In the above optional embodiment, by generating prompts based on the digital anchor's memory data of the target user, the generated prompts and interactive information are input into the large language model corresponding to the digital anchor, and the large language model is used to generate interactive reply information based on the prompts and interactive information. This allows the digital anchor to generate reply information under the prompt of the memory data, thereby improving the pertinence and accuracy of the replies generated by the digital anchor.
[0091] In an optional embodiment, if Figure 4 As shown, the digital anchor interaction method may further include steps S400 to S402:
[0092] Step S400, when no memory data can be found, obtain the character setting data of the digital anchor.
[0093] Step S402, using a large language model to generate interactive response information based on the interactive information and character setting data.
[0094] It should be understood that the digital anchor can adjust or train the large language model based on the character setting data to obtain a large language model that conforms to the character setting of the digital anchor. When receiving the interactive information of the target user, if the digital anchor's memory data of the target user cannot be found, the interactive information of the target user is directly input into the large language model corresponding to the digital anchor, and the large language model will output interactive reply information that conforms to the character setting of the digital anchor. For example, if the character setting data of the digital anchor includes a two-dimensional style, the generated interactive reply information will be information in a two-dimensional style.
[0095] In steps S400 to S402, the large language model may not be adjusted or trained in advance according to the character setting data of the digital anchor, but the character setting data of the digital anchor may be obtained, and then the large language model may be used to generate interactive response information according to the interactive information and the character setting data of the digital anchor. For example, the character setting data of the digital anchor may be used as a prompt to generate interactive response information based on the interactive information of the target user. In this way, interactive response information that conforms to the character setting of the digital anchor may be generated, and a large language model may be shared with other digital anchors, thereby reducing the cost of implementation.
[0096] In an optional embodiment, if Figure 5 As shown, step S202 may further include steps S500 to S502:
[0097] Step S500, obtaining the character setting data of the digital anchor.
[0098] Step S502, using a large language model to generate interactive response information of a target user according to the interactive information, memory data and character setting data.
[0099] As in the embodiment corresponding to the aforementioned steps S400 to S402, the large language model may not be adjusted or trained in advance according to the character setting data of the digital anchor, but after the digital anchor's memory data of the target user is queried, the character setting data of the digital anchor is obtained, the interactive information, the memory data and the character setting data of the digital anchor are input into the large language model together, and the large language model is used to output interactive response information that conforms to the character setting of the digital anchor. For example, the memory data and the character setting data of the digital anchor are used together as prompts to generate interactive response information for the interactive information of the target user.
[0100] In the above optional embodiment, by obtaining the character setting data of the digital anchor, a large language model is used to generate interactive response information of the target user according to the interactive information, memory data and the character setting data of the digital anchor. This can not only generate interactive response information that conforms to the character setting of the digital anchor, but also share a large language model with other digital anchors, thereby reducing the cost of implementation.
[0101] In an optional embodiment, if Figure 6 As shown, the digital anchor interaction method may further include steps S300 to S306:
[0102] Step S300, the interactive information and the interactive reply information constitute a complete conversation and are included in the temporary memory data to obtain updated temporary memory data.
[0103] Step S302, determining whether the number of updated temporary memory data reaches a preset number.
[0104] Step S304, when the number of updated temporary memory data reaches a preset number, new inductive memory data is generated according to the updated temporary memory data and the inductive memory data.
[0105] Step S306, updating the memory data according to the new summarized memory data.
[0106] In step S300, the interactive reply information is generated based on the interactive information and the memory data, so the interactive information and the interactive reply information constitute a complete conversation, that is, the interactive information and the interactive reply information generated based on the interactive information and the memory data constitute a complete conversation. Specifically, a similar structure such as "user question content" and "host reply content" can be used to constitute a complete conversation, and then the conversation is included in the temporary memory data. It can be understood that each time a conversation is included in the temporary memory data, the number of temporary memory data increases by one.
[0107] It should be noted that in the embodiment in which the aforementioned memory data includes the inductive memory data and the temporary memory data, the inductive memory data is generated according to a preset number of historical interaction data. When the historical interaction data reaches the preset number, one inductive memory data is generated according to the preset number of historical interaction data; when the new historical interaction data reaches the preset number again, another inductive memory data is generated according to the preset number of historical interaction data. That is, the number of inductive memory data may be more than one, but may be multiple.
[0108] In the embodiment corresponding to steps S300 to S306, when there is one piece of summarized memory data, if the number of updated temporary memory data reaches a preset number, then these updated temporary memory data and the existing summarized memory data are summarized and processed to obtain a new piece of summarized memory data, that is, the summarized memory data is one piece.
[0109] When the digital anchor's memory data of the target user is updated according to the new summarized memory data, it can be done after the digital anchor generates interactive reply information according to the interactive information and the memory data. Optionally, when the number of updated temporary memory data is less than a preset number, the digital anchor's memory data of the target user is updated according to the updated temporary memory data and the summarized memory data. At this time, the digital anchor's memory data of the target user includes data from two parts: the summarized memory data and the temporary memory data.
[0110] In the above optional embodiment, a complete conversation consisting of interactive information and interactive reply information is included in temporary memory data to obtain updated temporary memory data. When the number of updated temporary memory data reaches a preset number, new summarized memory data is generated based on the updated temporary memory data and the summarized memory data, and the memory data is updated based on the new summarized memory data. The memory data can be further summarized to obtain more concise and refined memory data, thereby further improving the efficiency of generating replies by the large language model corresponding to the digital anchor.
[0111] In an optional embodiment, if Figure 7 As shown, the digital anchor interaction method may further include steps S700 to S704:
[0112] Step S700, converting the interactive reply information into voice information.
[0113] Step S702, obtaining lip data corresponding to the digital anchor.
[0114] Step S704: Output voice information according to the lip shape data through the digital anchor.
[0115] Specifically, the server can convert the interactive response information in text form generated by the large language model into voice information using a speech synthesis method according to the corresponding voice of the digital anchor; obtain the lip shape data corresponding to each voice pre-constructed by the digital anchor, so that the digital anchor uses the corresponding virtual image to make the corresponding lip shape according to the lip shape data corresponding to the voice information while outputting the voice information, so that the audience can see the digital anchor use the corresponding virtual image to "read out" the interactive response information.
[0116] In the above optional embodiment, by converting the interactive reply information into voice information, obtaining the lip shape data corresponding to the digital anchor, and having the digital anchor output the voice information according to the lip shape data, the digital anchor can "read out" the corresponding interactive reply information, thereby enriching the interactive form of the digital anchor and improving the user experience.
[0117] In order to make this application easier to understand, the following Figure 8 An exemplary application is provided.
[0118] like Figure 8 As shown, digital anchor interaction can include the following:
[0119] 1. The audience enters the live broadcast room to watch the live broadcast of the digital anchor;
[0120] 2. The audience interacts with the digital anchor;
[0121] 3. Check whether there is any memory data of the digital anchor for the viewer. The memory data can be obtained based on historical interaction data such as room entry messages, historical bullet comments, and gift information;
[0122] 4. If the digital host has memory data about the viewer, a large language model is used to generate a corresponding reply to the viewer based on the memory data and the current interactive information.
[0123] Embodiment 2
[0124] Fig. 9 The block diagram of the digital anchor interactive device 800 according to the second embodiment of the present application is schematically shown. The device can be divided into one or more program modules, one or more program modules are stored in a storage medium, and are executed by one or more processors to complete the embodiment of the present application. The program module referred to in the embodiment of the present application refers to a series of computer program instruction segments that can complete specific functions. The following description will specifically introduce the functions of each program module in this embodiment. Fig. 9 As shown, the apparatus 800 may include: a query module 810 and a reply module 820, wherein:
[0125] A query module 810 is used to query whether there is memory data of the target user by the digital anchor when receiving the interaction information of the target user, wherein the memory data is obtained according to the historical interaction data between the target user and the digital anchor;
[0126] The reply module 820 is used to generate interactive reply information of the target user according to the interactive information and the memory data by using the large language model corresponding to the digital anchor when the memory data exists.
[0127] In an optional embodiment, the memory data includes inductive memory data and temporary memory data, the inductive memory data is generated based on a preset amount of historical interaction data, and the temporary memory data is the remaining historical interaction data except the inductive memory data.
[0128] In an optional embodiment, the device 800 is further used for:
[0129] The interactive information and the interactive reply information constitute a complete conversation and are included in the temporary memory data to obtain updated temporary memory data;
[0130] Determining whether the number of the updated temporary memory data reaches the preset number;
[0131] When the number of the updated temporary memory data reaches the preset number, generating new inductive memory data according to the updated temporary memory data and the inductive memory data;
[0132] The memory data is updated according to the new summarized memory data.
[0133] In an optional embodiment, the device 800 is further used for:
[0134] When the number of the updated temporary memory data is less than the preset number, the memory data is updated according to the updated temporary memory data and the summarized memory data.
[0135] In an optional embodiment, the device 800 is further used for:
[0136] If the memory data cannot be found, obtaining the character setting data of the digital anchor;
[0137] The interactive response information is generated by using the large language model according to the interactive information and the character setting data.
[0138] In an optional embodiment, the reply module 820 is further configured to:
[0139] Obtaining character setting data of the digital anchor;
[0140] The large language model is used to generate interactive response information of the target user according to the interactive information, the memory data and the character setting data.
[0141] In an optional embodiment, the reply module 820 is further configured to:
[0142] generating a prompt based on the memory data;
[0143] The prompt and the interaction information are input into the large language model, so as to utilize the large language model to generate the interaction response information according to the prompt and the interaction information.
[0144] In an optional embodiment, the device 800 is further used for:
[0145] Converting the interactive response information into voice information;
[0146] Obtaining lip shape data corresponding to the digital anchor;
[0147] The voice information is outputted by the digital anchor according to the lip shape data.
[0148] Embodiment 3
[0149] Fig.10The schematic diagram of the hardware architecture of a computer device 10000 suitable for implementing the digital anchor interaction method according to the third embodiment of the present application is shown schematically. In some embodiments, the computer device 10000 may be a terminal device such as a smart phone, a wearable device, a tablet computer, a personal computer, a vehicle terminal, a game console, a virtual device, a workbench, a digital assistant, a set-top box, a robot, etc. In other embodiments, the computer device 10000 may be a rack server, a blade server, a tower server, or a cabinet server (including an independent server, or a server cluster composed of multiple servers), etc. Fig.10 As shown, the computer device 10000 includes but is not limited to: a memory 10010, a processor 10020, and a network interface 10030 that can communicate with each other through a system bus. Among them:
[0150] The memory 10010 includes at least one type of computer-readable storage medium, and the readable storage medium includes a flash memory, a hard disk, a multimedia card, a card-type memory (such as an SD or DX memory), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 10010 may be an internal storage module of the computer device 10000, such as a hard disk or a memory of the computer device 10000. In other embodiments, the memory 10010 may also be an external storage device of the computer device 10000, such as a plug-in hard disk equipped on the computer device 10000, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Of course, the memory 10010 may also include both the internal storage module of the computer device 10000 and its external storage device. In this embodiment, the memory 10010 is generally used to store the operating system and various application software installed on the computer device 10000, such as the program code of the digital anchor interaction method, etc. In addition, the memory 10010 can also be used to temporarily store various data that have been output or are to be output.
[0151] In some embodiments, the processor 10020 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other chips. The processor 10020 is generally used to control the overall operation of the computer device 10000, such as performing control and processing related to data interaction or communication with the computer device 10000. In this embodiment, the processor 10020 is used to run the program code stored in the memory 10010 or process data.
[0152] The network interface 10030 may include a wireless network interface or a wired network interface, and the network interface 10030 is generally used to establish a communication link between the computer device 10000 and other computer devices. For example, the network interface 10030 is used to connect the computer device 10000 to an external terminal through a network, and to establish a data transmission channel and a communication link between the computer device 10000 and the external terminal. The network may be a wireless or wired network such as an intranet, the Internet, the Global System of Mobile communication (GSM), Wideband Code Division Multiple Access (WCDMA), 4G network, 5G network, Bluetooth, Wi-Fi, etc.
[0153] It should be pointed out that Fig.10 Only a computer device having components 10010 - 10030 is shown, but it should be understood that implementing all of the components shown is not a requirement, and more or fewer components may alternatively be implemented.
[0154] In this embodiment, the digital anchor interaction method stored in the memory 10010 can also be divided into one or more program modules and executed by one or more processors (such as processor 10020) to complete the embodiment of the present application.
[0155] Embodiment 4
[0156] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of the digital anchor interaction method in the embodiment are implemented.
[0157] In this embodiment, the computer-readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEP ROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the computer-readable storage medium may be an internal storage unit of a computer device, such as a hard disk or memory of the computer device. In other embodiments, the computer-readable storage medium may also be an external storage device of a computer device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device. Of course, the computer-readable storage medium may also include both the internal storage unit of the computer device and its external storage device. In this embodiment, the computer-readable storage medium is generally used to store an operating system and various application software installed on the computer device, such as the program code of the digital anchor interactive method in the embodiment. In addition, the computer-readable storage medium may also be used to temporarily store various types of data that have been output or are to be output.
[0158] Embodiment 5
[0159] An embodiment of the present application also provides a computer program product, including a computer program, which implements the method in the above embodiment when executed by a processor.
[0160] Obviously, those skilled in the art should understand that the modules or steps of the above-mentioned embodiments of the present application can be implemented by general-purpose computer devices, they can be concentrated on a single computer device, or distributed on a network composed of multiple computer devices, optionally, they can be implemented by executable program codes of computer devices, so that they can be stored in a storage device and executed by the computer device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. In this way, the embodiments of the present application are not limited to any specific combination of hardware and software.
[0161] It should be noted that the above are only preferred embodiments of the present application, and the patent protection scope of the present application is not limited thereto. Any equivalent structure or equivalent process transformation made using the contents of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.
Claims
1. A digital anchor interaction method, characterized in that: The method comprises: When receiving the interaction information of the target user, query whether there is memory data of the target user by the digital anchor, wherein the memory data is obtained according to the historical interaction data between the target user and the digital anchor; In the case where the memory data exists, the large language model corresponding to the digital anchor is used to generate interactive reply information of the target user according to the interactive information and the memory data.
2. The method according to claim 1, characterized in that The memory data includes inductive memory data and temporary memory data. The inductive memory data is generated according to a preset amount of historical interaction data, and the temporary memory data is the remaining historical interaction data except the inductive memory data.
3. The method according to claim 2, characterized in that The method further comprises: The interactive information and the interactive reply information constitute a complete conversation and are included in the temporary memory data to obtain updated temporary memory data; Determining whether the number of the updated temporary memory data reaches the preset number; When the number of the updated temporary memory data reaches the preset number, generating new inductive memory data according to the updated temporary memory data and the inductive memory data; The memory data is updated according to the new summarized memory data.
4. The method according to claim 3, characterized in that The method further comprises: When the number of the updated temporary memory data is less than the preset number, the memory data is updated according to the updated temporary memory data and the summarized memory data.
5. The method according to claim 1, characterized in that The method further comprises: If the memory data cannot be found, obtaining the character setting data of the digital anchor; The interactive response information is generated by using the large language model according to the interactive information and the character setting data.
6. The method according to claim 1, characterized in that The step of using the large language model corresponding to the digital anchor to generate the interactive reply information of the target user according to the interactive information and the memory data includes: Obtaining character setting data of the digital anchor; The large language model is used to generate interactive response information of the target user according to the interactive information, the memory data and the character setting data.
7. The method according to any one of claims 1 to 6, characterized in that: The step of using the large language model corresponding to the digital anchor to generate the interactive reply information of the target user according to the interactive information and the memory data includes: generating a prompt based on the memory data; The prompt and the interaction information are input into the large language model, so as to utilize the large language model to generate the interaction response information according to the prompt and the interaction information.
8. The method according to any one of claims 1 to 6, characterized in that: The method further comprises: Converting the interactive response information into voice information; Obtaining lip shape data corresponding to the digital anchor; The voice information is outputted by the digital anchor according to the lip shape data.
9. A digital anchor interactive device, characterized in that: The device comprises: A query module, configured to query whether there is memory data of the target user by the digital anchor when receiving the interaction information of the target user, wherein the memory data is obtained according to the historical interaction data between the target user and the digital anchor; A reply module is used to generate interactive reply information of the target user according to the interactive information and the memory data by using the large language model corresponding to the digital anchor when the memory data exists.
10. A computer device, characterized in that: include: at least one processor; and a memory communicatively connected to the at least one processor; wherein: The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.
11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the method according to any one of claims 1 to 8 is implemented.
12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.
Citation Information
Cited By
Data processing method and device
CN119544685A
Intelligent live broadcast interaction method and system based on AI virtual human
CN120786085A
