Auxiliary live broadcast processing method, device and electronic equipment

Abstract

The embodiments of the present invention provide a method, device and electronic device for processing auxiliary live broadcast, the method comprising triggering the display of an auxiliary live broadcast entrance according to the user's comment information and / or user behavior in the live broadcast room; performing user intent recognition in response to the user's input content in the auxiliary live broadcast interface, and obtaining recommended content related to the currently live broadcast commodity and / or merchant according to the recognized user intent; determining the content display style according to the host attributes and / or merchant attributes, and displaying the recommended content to the user in the content display style. The embodiments of the present invention enable users to timely understand the information they need during the live broadcast process, and enable users to feel the personalized style of the host and / or merchant, helping the host to take care of all users in the live broadcast room during the live broadcast process, thereby improving user satisfaction with the live broadcast.

Description

Technical Field

[0001] The present application relates to a method, device and electronic equipment for assisting live broadcast processing, and belongs to the field of computer technology. Background Art

[0002] With the development of the e-commerce industry, online commodity transactions have occupied a large part of people's daily consumption. In order to better promote commodity transactions, many e-commerce platforms provide live broadcast services. Merchants or individuals who are sellers can promote or promote commodities by live broadcasting to users.

[0003] In the live broadcast room of the existing technology, it is difficult for the anchor to communicate with each user while introducing the product. Therefore, some users' consultation requests may be ignored, resulting in reduced user satisfaction, user loss or affected product sales results. Summary of the invention

[0004] The embodiments of the present invention provide a method, device and electronic device for assisting live broadcast processing, so as to improve the live broadcast effect of a live broadcast room.

[0005] In order to achieve the above object, an embodiment of the present invention provides a method for processing auxiliary live broadcast, including:

[0006] Trigger the display of auxiliary live broadcast entrance according to the user's comment information and / or user behavior in the live broadcast room;

[0007] In response to the user's input content in the auxiliary live broadcast interface, the user's intention is identified, and according to the identified user's intention, recommended content related to the currently live broadcast product and / or merchant is obtained;

[0008] The content display style is determined according to the attributes of the live broadcast room, and the recommended content is displayed to the user in the content display style.

[0009] The embodiment of the present invention further provides a processing device for auxiliary live broadcast, including:

[0010] An entry trigger module is used to trigger the display of the auxiliary live broadcast entry according to the user's comment information and / or user behavior in the live broadcast room;

[0011] A content recommendation module is used to identify the user's intention in response to the user's input content in the auxiliary live broadcast interface, and obtain recommended content related to the currently live broadcast product and / or merchant according to the identified user intention;

[0012] The content display module is used to determine the content display style according to the attributes of the live broadcast room, and display the recommended content to the user in the content display style.

[0013] The embodiment of the present invention also provides a method for processing auxiliary live broadcast, including:

[0014] Obtain user input content and / or user behavior data during live broadcast, and identify user intent;

[0015] Obtain recommended content based on the identified user intent information;

[0016] The content display style is determined according to the attributes of the live broadcast room, and the recommended content is displayed to the user in the content display style.

[0017] The embodiment of the present invention further provides a processing device for auxiliary live broadcast, including:

[0018] The intention recognition module is used to obtain user input content and / or user behavior data during the live broadcast and perform user intention recognition;

[0019] A content recommendation module is used to obtain recommended content based on the identified user intent information;

[0020] The content display module is used to determine the content display style according to the attributes of the live broadcast room, and display the recommended content to the user in the content display style.

[0021] The embodiment of the present invention also provides a method for processing auxiliary live broadcast, including:

[0022] In response to a user's query request, creating a conversation space between the virtual anchor and the user;

[0023] Acquire recommended content according to the query request;

[0024] In the conversation space, the virtual anchor presents the recommended content to the user in a live broadcast through a one-to-one conversation mode, and continues to obtain and present the recommended content based on the user's comment information on the presented recommended content, and repeats the above actions until the user stops replying to information or a reply message to terminate the conversation is detected.

[0025] The embodiment of the present invention also provides a method for processing auxiliary live broadcast, including:

[0026] In response to a user's query request in the live broadcast room, entering a virtual display interface;

[0027] Acquire recommended content according to the query request;

[0028] In the virtual display interface, the associated virtual models and / or virtual models related to the recommended content are displayed, triggering the virtual anchor to introduce the recommended content in combination with the virtual models and / or virtual models.

[0029] An embodiment of the present invention further provides an electronic device, including:

[0030] Memory, used to store programs;

[0031] The processor is used to run the program stored in the memory to execute the above-mentioned auxiliary live broadcast processing method.

[0032] The embodiment of the present invention adopts an auxiliary live broadcast method to intelligently identify user intentions, recommend relevant display content to users based on user intentions, and provide a content display method with the characteristic style of the anchor and the merchant, so that users can timely understand the information they need during the live broadcast process, and can let users feel the personalized style of the anchor and / or the merchant, helping the anchor to take care of all users in the live broadcast room during the live broadcast process, thereby improving user satisfaction with the live broadcast.

[0033] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic diagram of the structure of an auxiliary live broadcast processing system according to an embodiment of the present invention;

[0035] Figure 2 This is one of the schematic diagrams of the live broadcast interaction interface in the auxiliary live broadcast scenario according to an embodiment of the present invention;

[0036] Figure 3 A schematic diagram of the logic of changes in the entrance text and entrance image of the virtual assistant according to an embodiment of the present invention;

[0037] Figure 4 This is a second schematic diagram of an interactive interface in an auxiliary live broadcast scenario according to an embodiment of the present invention;

[0038] Figure 5 This is a third schematic diagram of an interactive interface in an auxiliary live broadcast scenario according to an embodiment of the present invention;

[0039] Figure 6 This is a fourth schematic diagram of the interactive interface in the auxiliary live broadcast scenario according to an embodiment of the present invention;

[0040] Figure 7 This is a fifth schematic diagram of an interactive interface in an auxiliary live broadcast scenario according to an embodiment of the present invention;

[0041] Figure 8 This is a flow chart of a method for processing auxiliary live broadcast according to an embodiment of the present invention;

[0042] Fig. 9 This is one of the structural schematic diagrams of the auxiliary live broadcast processing device according to an embodiment of the present invention;

[0043] Fig.10 The second flowchart of the auxiliary live broadcast processing method according to the embodiment of the present invention;

[0044] Fig.11 This is a second structural diagram of the auxiliary live broadcast processing device according to an embodiment of the present invention;

[0045] Fig.12 The third flowchart of the auxiliary live broadcast processing method according to the embodiment of the present invention;

[0046] Fig.13 The fourth flowchart of the auxiliary live broadcast processing method according to the embodiment of the present invention;

[0047] Fig.14 FIG. 4 is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0049] In view of the problem in the prior art that the host is unable to take care of multiple users in the live broadcast room during the live broadcast, an embodiment of the present invention proposes a processing method for auxiliary live broadcast, which uses a virtual host to perform one-to-one interaction with users participating in the live broadcast. On the one hand, the user's intention is identified according to the user's input content and / or user behavior in the live broadcast room, and the push content related to the live broadcast merchant and / or product is intelligently pushed to the user, so that the user can obtain the information he wants to know in time. On the other hand, in terms of content display, a display style matching the host and / or merchant is adopted, so that the user can feel the personalized style of the host and / or merchant, thereby enhancing the user's good experience of the live broadcast.

[0050] like Figure 1As shown, it is a structural diagram of the auxiliary live broadcast processing system of an embodiment of the present invention. The system can be deployed on an e-commerce platform, and merchants can broadcast live based on the e-commerce platform. Users can use applications or web pages installed on terminal devices to access the live broadcast room and participate in the live broadcast. The terminal device can obtain user input content and / or user behavior data in real time and report it to the e-commerce platform. The e-commerce platform processes it through the auxiliary live broadcast processing system to form content displayed to the user, and finally displays the content in a display style that matches the anchor and / or merchant in the current live broadcast room. Among them, the user input content may include various comment information of the user in the live broadcast room and various interactive information entering the one-to-one mode through a designated entrance, and the user behavior may include various operation behaviors and silent behaviors used in the live broadcast room and the one-to-one mode. The auxiliary live broadcast processing system captures and intelligently identifies this information or behavior to present content to users in various ways. The core part of the auxiliary live broadcast processing system can be constructed through a machine learning model, which mainly includes an intent recognition model, a content recommendation model, and a content display model.

[0051] The input of the intention recognition model can be user input content and / or user behavior data, wherein the intention recognition for the input content can be recognized by a semantic recognition model based on natural language processing (NLP) technology, and the recognized semantic results are used as user intention information by inputting text content. In practical applications, the input content can be in the form of text, pictures, and sounds. Pictures and sounds can be converted into text input content by voice recognition models and / or image recognition models, and then the intention recognition based on the semantic recognition model is performed. The user input content can include not only the current input content but also the context content of the user in the live broadcast room and the previously recognized user intention, so as to have a comprehensive judgment and recognition of the user intention. For the intention recognition of user behavior data, a neural network model based on deep learning can be used to establish a mapping relationship between user behavior and user intention. The intention recognition model can be trained by using the massive user historical behavior data collected by the platform and the operation results finally performed by the user as training data to train the intention recognition model.

[0052] The user intention information identified by the intent recognition model will be input into the content recommendation model to perform content recommendation processing. The content recommendation model is associated with the content database, and the matching display content is obtained from the content database according to the user intention information. The content database may pre-store promotion activity data and / or product knowledge base data and / or merchant information data and / or basic question and answer data. The promotion activity data includes activity rules that are common to all products or merchants on the e-commerce platform, for example, the platform's unified promotion activity rules during the Double Eleven period. In addition, it may also include personalized rules for individual products or merchants. These activity rules can be set by the merchant themselves, for example, on the basis of a unified discount, further discounts are given to some products. Basic question and answer data may include general user rights protection content and explanations of basic processes such as transaction payment, return and exchange, etc. Product knowledge base data and merchant information data can be obtained based on the data accumulation of the e-commerce platform itself or from the knowledge base of other platforms on the network. The content recommendation model can be built based on a deep learning model to build a recommendation engine. By learning and mining massive data, the relationship between user intent and content data in the knowledge base is established. In the live broadcast process, according to the identified user intent, suitable content is matched for recommendation. In the process of content recommendation, in addition to user intent as the input of the model, the merchant information and product information of the current live broadcast are also used as model input, so that the obtained recommended content has a closer correlation with the merchant and product of the current live broadcast. Recommended content can include text content, picture content, video content, etc.

[0053] After the recommended content is determined, the output display style is determined by the content display model, and then the final output operation to the user is performed. In an embodiment of the present invention, a variety of methods can be used to display content to users, and the display methods can include voice playback, text reply, and image or video display. In order to allow users to feel the effect of one-to-one communication with the anchor, the display style can adopt a voice style, text style, and image rendering style with personalized characteristics of the anchor or merchant. For example, if the current anchor is an anchor with a certain degree of popularity, then the voice style can simulate the voice and tone of the anchor to play the content to be displayed. For another example, if the brand operated by the merchant belongs to a more fashionable brand, then in the image rendering style, a more avant-garde rendering mode can be adopted. For another example, even if the recommended content is displayed to the user in the form of a text reply, the tone of the anchor can be imitated in the language expression style. The above-mentioned voice style, text style, and image rendering style can be realized by a voice conversion model, a text conversion model, and an image rendering model, and these models can be trained by content data corresponding to the anchor or merchant. For example, in the voice conversion model, the voice data of a specific anchor (such as a certain internet celebrity anchor) or a specific type of anchor (such as a young woman or a middle-aged man, etc.) is used to train the model, so that after the voice model converts the input text, it has the voice style of the specific anchor or the specific type of anchor. Similarly, the text conversion model can also be trained with corpus corresponding to the specific anchor or the specific type of anchor, so that the text conversion model can convert the input text into a reply message with the text expression style of the specific anchor or the specific type of anchor. For the image rendering model, the images or videos provided by the merchant can also be used as training data, so that the image rendering model can render the recommended content in the form of images or videos to form a style corresponding to the merchant. It should be noted that although the voice style and text expression style generally correspond to the character, the merchant can be given anthropomorphic attributes by setting the attributes of the merchant. For example, if a merchant is set to the attributes of a teenager, then its voice and text expression style can adopt the style corresponding to the teenager. Similarly, the image rendering style can also be achieved by setting the attributes of the anchor. Therefore, in Figure 1 In the system structure shown, the preset live broadcast room attributes and recommended content can be used as inputs of the content display model, and then the corresponding model is determined according to the live broadcast room attributes to perform content display processing. The live broadcast room attributes mentioned here can include anchor attributes and / or merchant attributes, where anchor attributes belong to personal style attributes, while merchant attributes focus on brand style attributes. In addition, special attributes can also be set for the live broadcast room, such as a live broadcast room with a magical color, the content display style of the live broadcast room has a magical display style, and the rendering style of the image adds magical elements, etc.

[0054] The above introduces the basic technical principles of the auxiliary live broadcast processing system. The following introduces an example of the specific interface form of the auxiliary live broadcast of the embodiment of the present invention. The auxiliary live broadcast of the embodiment of the present invention can be implemented by means of a virtual assistant, that is, the user interacts with the virtual assistant to achieve the effect of one-to-one interaction with the anchor to perform auxiliary live broadcast processing. Figure 2 As shown, it is one of the schematic diagrams of the live broadcast interaction interface in the assisted live broadcast scenario of an embodiment of the present invention. The user can enter the interface for interacting with the virtual assistant by clicking the "little assistant" icon on the live broadcast interface. The "little assistant" icon can be regarded as an auxiliary live broadcast entrance.

[0055] The display form of the auxiliary live broadcast entrance can be triggered by the user's comment information and / or user behavior in the live broadcast room, so as to attract users to interact with the virtual anchor. The specific method may include: detecting keywords in the comment information, and if the keywords hit the preset question, then according to the preset question, determine the corresponding entrance copy and / or entrance image, and display the auxiliary live broadcast entrance. Figure 2 In the live broadcast interface shown in the left picture, the auxiliary live broadcast entrance is in the default state, that is, the icon of the "little assistant" is in the default state. When the system detects that the user comments on the dress product being broadcast live, the virtual assistant is triggered to enter the trigger guidance state, such as Figure 2 As shown in the middle picture, the icon of the "little assistant" changes to an image of a hand-raising state, and the entry copy "click me for a dress" is displayed below it. In addition, various user behaviors can also be detected to present different entry display states. For example, you can pay attention to the user's silence time in the live broadcast room. User silence is regarded as a user behavior. If the user is silent for too long, it may mean that the user is not interested in the content of the current live broadcast. In this case, you can actively change the entry image to attract the user's attention. Specifically, you can detect the user's silence time in the live broadcast room. If the silence time is greater than the preset time threshold, the auxiliary live broadcast entrance is displayed with the corresponding entry copy and / or entry image. For example, when the user has been silent for a long time, you can actively change the "little assistant" icon to an image of a hand-raising state, and cooperate with the entry copy "click me for a discount". In addition, if the user's silence behavior occurs after the virtual assistant is triggered and guided, you can use different entry images and entry copy. For example, after the entry copy "click me for a dress" is displayed, the user ignores it, you can Figure 2 As shown in the right picture, the icon of "little assistant" is changed into a sad and tearful image, and the entrance copy of "ignore me" is added to further attract users' attention. Figure 3 As shown, it is a schematic diagram of the change logic of the entrance copy and entrance image of the virtual assistant in an embodiment of the present invention, as an example, Figure 3The figure shows the changes in the entry copy and image of the virtual assistant in some scenarios. Generally speaking, a variety of entry copy will be triggered based on user comments to guess the issues that users care about and look forward to further communication with users. These questions may include size, discounts, purchase guides, etc., which all depend on the analysis and judgment of user comment information. After the entry copy in the middle part of Figure 3 is displayed, if the user does not click to trigger the one-to-one state, for example, if it is not triggered within 5 seconds, it will turn to the image state of "ignore me" and then return to the default assistant image. In addition, if the user is silent for more than 20 seconds in the live broadcast room, the entry copy "Click me for discounts" that provides users with discount information will be triggered.

[0056] Furthermore, if the user clicks Figure 2 Click the virtual assistant icon in the video to enter the one-to-one communication state of assisting the live broadcast. Figure 4 As shown, it is a second schematic diagram of the interactive interface in the auxiliary live broadcast scenario of an embodiment of the present invention. Figure 4 The interface shown is that the user has entered a one-to-one interaction state with the virtual assistant. As shown in the figure, after entering the chat interface, Figure 4 As shown in the left picture, the virtual assistant will actively greet the user with a specific image and text, and if the user does not respond, it can also actively display relevant topic options in the auxiliary live broadcast interface based on the current live broadcast process and / or promotional activities, such as Figure 4 The middle picture shows that some frequently-watched topics are shown to users based on the current promotional activities to guide user interaction. In addition, during the interaction between the virtual assistant and the user, if the user's intention is not clearly recognized, some example questions can be given or guessed at the questions the user wants to ask, etc., and presented in the interactive interface to enable the interaction to continue. In addition, if Figure 5 As shown, it is the third interactive interface diagram in the auxiliary live broadcasting scenario of the embodiment of the present invention, and the user can input information by voice or by text in the dialog box. Figure 6 and Figure 7 As shown, it is the fourth and fifth interactive interface schematic diagrams in the auxiliary live broadcast scenario of an embodiment of the present invention, which show an example of the auxiliary live broadcast interactive interface presenting other content. Figure 6 In the video, the virtual assistant shows the user relevant content about the discount. Figure 7 In the , the virtual assistant presents the user with descriptions of single items and multiple items.

[0057] The above describes the process of the virtual anchor helping the anchor to interact with the users in the live broadcast room in an auxiliary form. In fact, the virtual anchor can also collaborate with the real anchor. The virtual anchor cooperates with the real anchor to display various content according to the current live broadcast process. For example, the real anchor is selling a pair of sports shoes live. When introducing the main functions and characteristics of the sports shoes, the virtual anchor can intervene in the live broadcast and introduce the functions and characteristics of each part of the sports shoes to the user in the live broadcast room in detail in the form of 3D images and text and voice. The virtual anchor can trigger the joining of the collaborative live broadcast by intelligently identifying the anchor content of the real anchor and / or the comments of the users in the live broadcast room. The real anchor can trigger the virtual anchor to join in the form of voice or text. For example, the real anchor can say "please let "XX" introduce the specific characteristics of the sports shoes", thereby triggering the virtual anchor to join, where "XX" can be the name of the preset virtual anchor. For another example, when more than a predetermined number of users reply to messages in the live broadcast room and wish to learn more about the product details, the virtual anchor can be triggered to give a detailed introduction. The virtual anchor's presentation content can be based on real-time identified user needs, and recommended content can be obtained from the database and presented live.

[0058] In addition, the virtual anchor can appear in the virtual live broadcast room with a preset virtual image. For example, in order to match with the real anchor, the virtual anchor can be presented by processing the real anchor by computer to obtain a virtual image. This virtual image has some image characteristics of the real anchor and the style of the virtual image. For example, the image of the real anchor is used as the basis to generate a virtual two-dimensional image. The images of the virtual anchor and the real anchor can be presented in a rotating form. When the real anchor is broadcasting live, the image of the real anchor is temporarily displayed, and when the virtual anchor needs to intervene in the live broadcast, the image of the virtual anchor is displayed. In addition, the real anchor and the virtual anchor can be presented together in the live broadcast room, thereby presenting the state of the co-anchor. In addition, in order to make users feel more involved in the live broadcast, when the user speaks, a virtual dialogue image corresponding to the user can also be generated and displayed in the live broadcast room, so as to make the live broadcast effect richer and give users a stronger sense of participation.

[0059] It can be seen from the above examples that the embodiments of the present invention use an assisted live broadcast method to intelligently identify user intentions, recommend relevant display content to users based on user intentions, and provide a content display method with the characteristic style of the anchor and the merchant, so that users can promptly understand the information they need during the live broadcast, and can let users feel the personalized style of the anchor and / or the merchant, helping the anchor to take care of all users in the live broadcast room during the live broadcast, thereby improving user satisfaction with the live broadcast.

[0060] The technical solution of the present invention is further illustrated by some specific embodiments below.

[0061] Embodiment 1

[0062] like Figure 8 As shown, it is one of the flow diagrams of the processing method for auxiliary live broadcasting according to an embodiment of the present invention. The method can be applied to an e-commerce platform that provides e-commerce services to merchants and consumers. A user accesses the e-commerce platform through an application or webpage installed on a terminal device and enters a live broadcast room to participate in the live broadcasting. The method may include:

[0063] S101: Trigger the display of the auxiliary live broadcast entrance according to the user's comment information and / or user behavior in the live broadcast room. The comment information can be various information sent by the user in the public chat area of ​​the live broadcast room. By detecting the keywords in the comment information, the questions that the user may be interested in can be determined. Specifically, the detected keywords can be matched with some preset questions, such as price questions, size questions, color questions, etc. If the keywords hit the preset questions, the corresponding first entrance copy and / or first entrance image are determined according to the preset questions, and the auxiliary live broadcast entrance is displayed. The auxiliary live broadcast entrance can be set at any position on the live broadcast room interface, and can be specifically in the form of an icon plus copy. For example, the front Figure 2 As shown in the example, a cartoon character avatar combined with text is used as an icon of the auxiliary live broadcast entrance. The entrance image and entrance copy can change according to the hit question. In addition, the user behavior of the user in the live broadcast room may include various operations of the user on the live broadcast room interface, such as scrolling, clicking or viewing various pictures and links in the live broadcast, etc., and may also include silent behavior without any operation for a long time. By judging these user behaviors, the user's intention can also be inferred and the auxiliary live broadcast entrance can be triggered to display. For example, if the user repeatedly views the pictures of sportswear released in the live broadcast room, the auxiliary live broadcast entrance that displays the entrance copy and entrance image related to sportswear can be triggered. In addition, during the live broadcast process, some users often remain silent for a long time. It may be that the user has lost interest in the live broadcast or the question he wants to know is not answered by the anchor at present. In view of this situation, the auxiliary live broadcast entrance can be actively activated to attract users for one-to-one communication. Specifically, the user's silence time in the live broadcast room can be detected. If the silence time is greater than the preset time threshold, the auxiliary live broadcast entrance is displayed with the preset second entrance copy and / or second entrance image. For example, the copy of the promotional activities is displayed to the user, and a more active entrance image is used to attract the user's attention. The user can enter the auxiliary live broadcast interface through the auxiliary live broadcast entrance by clicking, etc., so that Figure 4 and Figure 5 The interface shown is for one-on-one interaction with a virtual assistant.

[0064] S102: In response to the user's input content in the auxiliary live broadcast interface, the user's intention is identified, and the recommended content related to the currently live broadcast product and / or merchant is obtained according to the identified user intention. When the user enters the auxiliary live broadcast interface, the user's input content in the interactive interface will be sent to the e-commerce platform, and the user's intention is identified. After the user's intention is identified, the relevant recommended content can be obtained from the content recommendation database of the e-commerce platform according to the user's intention based on the recommendation model set on the e-commerce platform. Specifically, the user's intention information and the currently live product information and / or merchant information can be used as a recommendation model to obtain recommended content with a high degree of association with the currently live product and / or merchant. Among them, the data in the content recommendation library may include promotional activity data and / or product knowledge base data and / or merchant information data and / or basic question and answer data, etc. In addition, the data in the content recommendation database can come from the accumulation of the e-commerce platform itself, or from a third-party knowledge database, etc.

[0065] The above-mentioned intention recognition processing can be completed using a semantic recognition model. The semantic recognition model can perform semantic recognition based on the text content input by the user and generate user intention information in text form, which is then used as the input of the recommendation model. In the auxiliary live broadcast interface, the user's input content can be text input content, voice input content and / or image input content. For voice input content and image input content, the content can be recognized by the voice recognition model and the image recognition model respectively, and after converting it into text input content, the intention recognition processing is performed.

[0066] S103: Determine the content display style according to the live broadcast room attributes, and display the recommended content to the user in the content display style. In the embodiment of the present invention, in order to allow users to better experience the one-to-one interaction effect, the unique style of the anchor and / or the merchant can be integrated into the content display, and the recommended content can be displayed to the user in a variety of ways such as text, voice, image or video.

[0067] Specifically, for the voice display method, a voice conversion model corresponding to the attributes of the live broadcast room can be used to display the recommended content to the user in the form of voice playback in a voice style corresponding to the host and / or; for the text display method, a text conversion model corresponding to the attributes of the live broadcast room can be used to display the recommended content to the user in the form of text in a text style corresponding to the host and / or; for the image and video display method, an image rendering model corresponding to the attributes of the live broadcast room can be used to display the recommended content to the user in the form of images and / or videos in an image rendering style corresponding to the host and / or. The anchor attribute mentioned here may be information that identifies the anchor or the anchor type. For example, if the current anchor is a specific star, then the text conversion model, speech conversion model, and image rendering model corresponding to the star may be used to perform content display processing. These models are all trained using the training data corresponding to the star, so that these models have the star's unique style. For example, in speech processing, the star's speech data may be used as training data, so that the speech conversion model has the ability to imitate the star's tone and voice. In text processing, the corpus that the star usually uses may be used as training data, so that the text conversion model has the ability to imitate the star's writing style. In image rendering, based on artificially generated training data, for example, according to the characteristics of a star, the image rendering style may be set to a romantic style. Then, the image rendering model may be trained using data with this style, so that the image rendering model has the image rendering style corresponding to the star. In addition, the host's attributes can also be set to a certain type, such as a girl, a child, an old man, etc. Similar to the principle introduced earlier, it only needs to be trained with the corresponding training data to enable the text conversion model, the speech conversion model, and the image rendering model to have the ability to imitate the corresponding display style, thereby displaying content in a specific display style.

[0068] In addition, in the auxiliary live broadcast interface, relevant topic options can be displayed based on the current live broadcast progress and / or promotional activities, so that users have more topic choices and can cooperate with the live broadcast promotion in real time.

[0069] Furthermore, the direct interaction between the user and the virtual anchor may form multiple rounds of dialogue, and the virtual anchor continuously identifies the user's consulting needs and provides answers and replies in real time. Specifically, the above method may also include: determining the user's consulting question based on the user's reply information; obtaining the reply information corresponding to the consulting question, displaying the reply information to the user in the content display style, and repeating the above actions until the user stops replying to the information or the reply information for terminating the session is detected. The reply information for terminating the session mentioned here can be, for example, "Thank you", "I understand", "I have no questions", etc.

[0070] The assisted live broadcast processing method of the embodiment of the present invention adopts an assisted live broadcast method to intelligently identify user intentions, recommend relevant display content to users based on user intentions, and provide a content display method with the characteristic style of the anchor and the merchant, so that users can timely understand the information they need during the live broadcast process, and can let users feel the personalized style of the anchor and / or the merchant, helping the anchor to take care of all users in the live broadcast room during the live broadcast process, thereby improving user satisfaction with the live broadcast.

[0071] Embodiment 2

[0072] like Fig. 9 As shown, it is one of the structural schematic diagrams of the auxiliary live broadcast processing device of an embodiment of the present invention. The device can be applied to an e-commerce platform that provides e-commerce services to merchants and consumers. Users access the e-commerce platform through applications or web pages installed on terminal devices and enter the live broadcast room to participate in the live broadcast. The device may include:

[0073] The entry trigger module 11 is used to trigger the display of the auxiliary live broadcast entry according to the comment information and / or user behavior of the user in the live broadcast room. Among them, the comment information can be various information sent by the user in the public chat area of ​​the live broadcast room. By detecting the keywords in the comment information, the questions that the user may be interested in are judged. If the keywords hit the preset questions, the corresponding first entry copy and / or first entry image are determined according to the preset questions to display the auxiliary live broadcast entry. The auxiliary live broadcast entry can be set at any position on the live broadcast room interface, and can specifically be in the form of an icon plus copy. The entry image and the entry copy can change according to the hit question. In addition, the user behavior of the user in the live broadcast room can include various operations of the user on the live broadcast room interface, such as scrolling, clicking or viewing various pictures and links in the live broadcast, etc., and can also include silent behavior without any operation for a long time. By judging these user behaviors, the user intention can also be inferred and the display of the auxiliary live broadcast entry can be triggered. During the live broadcast process, some users often remain silent for a long time. This may be because the users have lost interest in the live broadcast or the host has not yet answered the questions they want to know. In this case, the auxiliary live broadcast entrance can be actively activated to attract users to communicate one-on-one. Specifically, the user's silence time in the live broadcast room can be detected. If the silence time is greater than the preset time threshold, the auxiliary live broadcast entrance is displayed with the preset second entrance copy and / or second entrance image. For example, the copy of the promotional activities is displayed to the user, and a more active entrance image is used to attract the user's attention. The user can enter the auxiliary live broadcast interface through the auxiliary live broadcast entrance by clicking or other methods.

[0074] The content recommendation module 12 is used to respond to the user's input content in the auxiliary live broadcast interface, identify the user's intention, and obtain recommended content related to the currently live broadcast goods and / or merchants according to the identified user intention. When the user enters the auxiliary live broadcast interface, the user's input content in the interactive interface will be sent to the e-commerce platform, and the user's intention will be identified. After identifying the user's intention, it is possible to obtain relevant recommended content from the content recommendation database of the e-commerce platform based on the recommendation model set on the e-commerce platform and the user's intention. Specifically, the user's intention information and the currently live broadcast product information and / or merchant information can be used as a recommendation model to obtain recommended content with a high degree of association with the currently live broadcast product and / or merchant. Among them, the data in the content recommendation library can include promotional activity data and / or product knowledge base data and / or merchant information data and / or basic question and answer data, etc. In addition, the data in the content recommendation database can come from the accumulation of the e-commerce platform itself, or from a third-party knowledge database, etc.

[0075] The above-mentioned intention recognition processing can be completed using a semantic recognition model. The semantic recognition model can perform semantic recognition based on the text content input by the user and generate user intention information in text form, which is then used as the input of the recommendation model. In the auxiliary live broadcast interface, the user's input content can be text input content, voice input content and / or image input content. For voice input content and image input content, the content can be recognized by the voice recognition model and the image recognition model respectively, and after converting it into text input content, the intention recognition processing is performed.

[0076] The content display module 13 is used to determine the content display style according to the live broadcast room attributes, and display the recommended content to the user in the content display style. The live broadcast room attributes mentioned here may include anchor attributes and / or merchant attributes, wherein the anchor attributes belong to personal style attributes, while the merchant attributes focus on brand style attributes. In an embodiment of the present invention, in order to allow users to better experience the one-to-one interaction effect, the unique style of the anchor and / or merchant can be integrated into the content display, and the recommended content can be displayed to the user in a variety of ways such as text, voice, image or video.

[0077] Specifically, for voice display method, a voice conversion model corresponding to the attributes of the live broadcast room can be used, for example, recommended content can be displayed to users in the form of voice playback in a voice style corresponding to the host and / or merchant; for text display method, a text conversion model corresponding to the attributes of the live broadcast room can be used, and recommended content can be displayed to users in the form of text in a text style corresponding to the host and / or merchant; for image and video display methods, an image rendering model corresponding to the attributes of the live broadcast room can be used, and recommended content can be displayed to users in the form of images and / or videos in an image rendering style corresponding to the host and / or merchant.

[0078] The auxiliary live broadcast processing device of the embodiment of the present invention adopts an auxiliary live broadcast method to intelligently identify user intentions, recommend relevant display content to users based on user intentions, and provide a content display method with the characteristic style of the anchor and the merchant, so that users can timely understand the information they need during the live broadcast, and can let users feel the personalized style of the anchor and / or the merchant, helping the anchor to take care of all users in the live broadcast room during the live broadcast, thereby improving user satisfaction with the live broadcast.

[0079] Embodiment 3

[0080] like Fig.10 As shown, it is a second flow diagram of the processing method for auxiliary live broadcasting according to an embodiment of the present invention. The method can be applied to an e-commerce platform that provides e-commerce services to merchants and consumers. A user accesses the e-commerce platform through an application or webpage installed on a terminal device and enters a live broadcast room to participate in the live broadcasting. The method may include:

[0081] S201: Obtain user input content and / or user behavior data during the live broadcast process, and perform user intent recognition. The user input content mentioned here can be various forms of content input by the user in the live broadcast room, including voice, text, pictures, and videos, etc. These contents may be comments on the current live broadcast content, or they may be questions asked to the anchor, etc. User behavior data can be various operating behaviors used in the live broadcast room, such as turning pages down, clicking various links, or being silent, etc. User input content and / or user behavior data can be uploaded to the e-commerce platform, and the user's intention is recognized through the intention recognition model set in the e-commerce platform to generate user intention information.

[0082] S202: Obtain recommended content based on the identified user intent information. A content recommendation model for recommending content based on user intent information may be set up on the e-commerce platform. The content recommendation model may be associated with a content database, and matching recommended content may be obtained from the content database. In addition, the content recommendation model may also obtain various recommended content from an authorized third-party platform via the Internet.

[0083] S203: Determine the content display style according to the live broadcast room attributes, and display the recommended content to the user in the content display style. The live broadcast room attributes may include the host attributes and / or the merchant attributes. In order to allow users to better experience the one-to-one interaction effect, the host and / or merchant's unique style can be integrated into the content display, and the recommended content can be displayed to the user in a variety of ways such as text, voice, image or video. Specifically, this can be achieved through a voice conversion model, a text conversion model, and an image rendering model corresponding to the live broadcast room attributes.

[0084] The specific description of the above-mentioned processing process, the detailed description of the technical principle and the detailed analysis of the technical effect have been described in detail in the previous embodiments and will not be repeated here.

[0085] The assisted live broadcast processing method of the embodiment of the present invention adopts an assisted live broadcast method to intelligently identify user intentions, recommend relevant display content to users based on user intentions, and provide a content display method with the characteristic style of the anchor and the merchant, so that users can timely understand the information they need during the live broadcast process, and can let users feel the personalized style of the anchor and / or the merchant, helping the anchor to take care of all users in the live broadcast room during the live broadcast process, thereby improving user satisfaction with the live broadcast.

[0086] Embodiment 4

[0087] like Fig.11 As shown, it is a second structural schematic diagram of the auxiliary live broadcast processing device of an embodiment of the present invention. The device can be applied to an e-commerce platform that provides e-commerce services to merchants and consumers. Users access the e-commerce platform through applications or web pages installed on terminal devices and enter the live broadcast room to participate in the live broadcast. The device may include:

[0088] The intention recognition module 21 is used to obtain user input content and / or user behavior data during the live broadcast process, and to recognize user intent. The user input content mentioned here can be various forms of content input by the user in the live broadcast room, including voice, text, pictures, and videos, etc. These contents may be comments on the current live broadcast content, or they may be questions asked to the anchor, etc. User behavior data can be various operating behaviors used in the live broadcast room, such as turning pages, clicking various links, or being silent. The intention recognition module 21 may include an intention recognition model set in the e-commerce platform, and the user input content and / or user behavior data can be uploaded to the e-commerce platform. The user's intention is recognized through the intention recognition model to generate user intent information.

[0089] The content recommendation module 22 is used to obtain recommended content based on the identified user intention information.

[0090] The content recommendation module 22 may include a content recommendation model set on the e-commerce platform, and the content recommendation model may be associated with a content database, and matching recommended content may be obtained from the content database. In addition, the content recommendation model may also obtain various recommended content from an authorized third-party platform through the Internet.

[0091] The content display module 23 is used to determine the content display style according to the attributes of the live broadcast room, and to display the recommended content to the user in the content display style. The live broadcast room attributes mentioned here may include anchor attributes and / or merchant attributes. In order to allow users to better experience the one-to-one interaction effect, the unique style of the anchor and / or merchant can be integrated into the content display, and the recommended content can be displayed to the user in a variety of ways such as text, voice, image or video. Specifically, the content display module 23 may include a voice conversion model, a text conversion model, and an image rendering model corresponding to the attributes of the live broadcast room, and these models are used to realize content display in a style corresponding to the anchor and / or merchant.

[0092] The specific description of the above-mentioned processing process, the detailed description of the technical principle and the detailed analysis of the technical effect have been described in detail in the previous embodiments and will not be repeated here.

[0093] The auxiliary live broadcast processing device of the embodiment of the present invention adopts an auxiliary live broadcast method to intelligently identify user intentions, recommend relevant display content to users based on user intentions, and provide a content display method with the characteristic style of the anchor and the merchant, so that users can timely understand the information they need during the live broadcast, and can let users feel the personalized style of the anchor and / or the merchant, helping the anchor to take care of all users in the live broadcast room during the live broadcast, thereby improving user satisfaction with the live broadcast.

[0094] Embodiment 5

[0095] The embodiment of the present invention also provides a method for processing auxiliary live broadcast, such as Fig.12 As shown, it is a third flow chart of the auxiliary live broadcast processing method according to an embodiment of the present invention, and the method includes:

[0096] S301: In response to a user's query request, a conversation space between the virtual anchor and the user is created. When a user makes some specific query requests in the live broadcast room, for example, the user asks "I want to learn more about this clothing" in the live broadcast room, a separate conversation space can be created between the virtual anchor and the user, thereby entering a one-to-one conversation state.

[0097] S302: Obtain recommended content according to the query request. Based on the user's query request, the e-commerce platform can query the background database for relevant recommended content.

[0098] S303: In the conversation space, the virtual host displays recommended content to the user in a live broadcast through a one-to-one conversation mode, and continues to obtain and display recommended content based on the user's comment information on the displayed recommended content, and repeats the above actions until the user stops replying to the information or a reply message to terminate the conversation is detected.

[0099] In the one-on-one conversation mode, users can have multiple rounds of conversations with the virtual anchor, thereby digging deep into the content that the user wants to know and continuously displaying content, so that users can have a full understanding of the products and related information, to make up for the fact that the real anchor cannot take care of every user during the live broadcast.

[0100] In addition, the embodiment of the present invention also provides another method for processing auxiliary live broadcast, such as Fig.13 As shown, it is a fourth flow chart of the auxiliary live broadcast processing method according to an embodiment of the present invention, and the method includes:

[0101] S401: In response to a user's query request in the live broadcast room, a virtual display interface is entered. The virtual display interface may be a one-to-one interface between a virtual anchor and a user, and a display space may be reserved in the interface for the user to display a 3D model.

[0102] S402: Obtaining recommended content according to the query request Based on the user's query request, the e-commerce platform can query the background database for relevant recommended content.

[0103] S403: In the virtual display interface, display the virtual models and / or virtual models associated with the recommended content, triggering the virtual anchor to introduce the recommended content in combination with the virtual models and / or virtual models. The virtual models and / or virtual models here can be in 3D form, and can perform functions such as perspective change and zooming. Furthermore, in the process of introducing the virtual models and / or virtual models, user interactive operations can also be introduced. Specifically, in response to the user's click operation on the virtual model and / or virtual model, the virtual anchor can be triggered to introduce the content of the part corresponding to the click operation. For example, if the user clicks on the pocket part of the clothes worn by the model, the view can be enlarged, and the virtual anchor will give a detailed introduction to the details and functions of the pocket.

[0104] In an embodiment of the present invention, by establishing a one-to-one conversation space or display space between a virtual anchor and a user based on the user's query request, the product details and other information that the user wants to know can be displayed to the user in more detail, thereby achieving a better live broadcast effect of the product.

[0105] Embodiment 6

[0106] The above embodiments describe the process of the auxiliary live broadcast processing method and the corresponding device structure. The functions of the above method and device can be implemented by means of an electronic device, such as Fig.14 As shown, it is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention, which specifically includes: a memory 110 and a processor 120.

[0107] The memory 110 is used to store programs.

[0108] In addition to the above-mentioned programs, the memory 110 may also be configured to store various other data to support operations on the electronic device. Examples of such data include instructions for any application or method operating on the electronic device, contact data, phone book data, messages, pictures, videos, etc.

[0109] The memory 110 may be implemented by any type of volatile or nonvolatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0110] The processor 120 is coupled to the memory 110 and is used to execute the program in the memory 110 to perform the operating steps of the auxiliary live broadcast processing method described in the above embodiments.

[0111] In addition, the processor 120 may also include the various modules described in the aforementioned embodiments to perform auxiliary live broadcast processing, and the memory 110 may be used, for example, to store data required for these modules to perform operations and / or output data.

[0112] The specific description of the above-mentioned processing process, the detailed description of the technical principle and the detailed analysis of the technical effect have been described in detail in the previous embodiments and will not be repeated here.

[0113] Furthermore, as shown in the figure, the electronic device may also include other components such as a communication component 130, a power component 140, an audio component 150, and a display 160. The figure only schematically shows some components, which does not mean that the electronic device only includes the components shown in the figure.

[0114] The communication component 130 is configured to facilitate wired or wireless communication between the electronic device and other devices. The electronic device can access a wireless network based on a communication standard, such as WiFi, 2G, 3G, 4G / LTE, 5G and other mobile communication networks, or a combination thereof. In an exemplary embodiment, the communication component 130 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 130 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0115] The power supply component 140 provides power to various components of the electronic device. The power supply component 140 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the electronic device.

[0116] The audio component 150 is configured to output and / or input audio signals. For example, the audio component 150 includes a microphone (MIC), and when the electronic device is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 110 or sent via the communication component 130. In some embodiments, the audio component 150 also includes a speaker for outputting audio signals.

[0117] The display 160 includes a screen, which may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation.

[0118] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for processing auxiliary live broadcast, comprising: Trigger the display of auxiliary live broadcast entrance according to the user's comment information and / or user behavior in the live broadcast room; In response to the user input content in the auxiliary live broadcast interface, the user intention is identified, and according to the identified user intention, the recommended content related to the currently live broadcast commodity and / or merchant is obtained; wherein the user intention is obtained by performing user intention identification on the user input content according to the semantic recognition model; The content display style is determined according to the attributes of the live broadcast room, and the recommended content is displayed to the user in the content display style. The content display style includes at least one of a voice style, a text style, and an image rendering style with personalized characteristics of the anchor or the merchant.

2. The method according to claim 1, wherein: According to the user's comment information and / or user behavior in the live broadcast room, the auxiliary live broadcast entrance is triggered to display: Detecting keywords in the comment information, and if the keywords match a preset question, determining a corresponding first entry copy and / or first entry image according to the preset question, and displaying an auxiliary live broadcast entry; and / or, Detect the user's silence time in the live broadcast room. If the silence time is greater than a preset time threshold, display the auxiliary live broadcast entrance with a preset second entrance copy and / or second entrance image.

3. The method according to claim 1, wherein: In response to the user's input content in the auxiliary live broadcast interface, the user's intention is identified, and according to the identified user's intention, the recommended content related to the currently live broadcast product and / or merchant is obtained, including: Use the semantic recognition model to identify the user's intention based on the user's input content and generate user intention information; The recommendation model is used to obtain the recommended content from a content database according to the user intention information and the currently broadcast commodity information and / or merchant information.

4. The method according to claim 3, wherein: Also includes: If the input content is voice input content and / or image input content, a voice recognition model and / or an image recognition model is used to convert the voice input content and / or the image input content into text input content, and then intent recognition is performed.

5. The method according to claim 3, wherein: The content database stores promotion activity data and / or commodity knowledge base data and / or merchant information data and / or basic question and answer data.

6. The method according to claim 1, wherein: Determining a content display style according to the live broadcast room attribute, and displaying the recommended content to the user in the content display style includes: Using a voice conversion model corresponding to the attributes of the live broadcast room, and using a voice style corresponding to the anchor and / or the host, the recommended content is presented to the user in the form of voice playback; and / or, Using a text conversion model corresponding to the attributes of the live broadcast room, displaying the recommended content to the user in text form in a text style corresponding to the anchor and / or; and / or, The recommended content is displayed to the user in the form of images and / or videos using an image rendering model corresponding to the attributes of the live broadcast room and an image rendering style corresponding to the anchor and / or.

7. The method according to claim 1, wherein: Also includes: In the auxiliary live broadcast interface, relevant topic options are displayed according to the current live broadcast progress and / or promotional activities.

8. The method according to claim 1, wherein: Also includes: Determine the user's consultation question based on the user's reply information; Acquire reply information corresponding to the consultation question, display the reply information to the user in the content display style, and repeat the above actions until the user stops replying to the information or a reply information terminating the session is detected.

9. A processing device for auxiliary live broadcast, comprising: An entry trigger module is used to trigger the display of the auxiliary live broadcast entry according to the user's comment information and / or user behavior in the live broadcast room; A content recommendation module is used to identify the user's intention in response to the user's input content in the auxiliary live broadcast interface, and obtain recommended content related to the currently live broadcast commodity and / or merchant according to the identified user intention; wherein the user intention is obtained by identifying the user's input content according to the semantic recognition model; The content display module is used to determine the content display style according to the attributes of the live broadcast room, and display the recommended content to the user in the content display style. The content display style includes at least one of the voice style, text style and image rendering style with the personalized characteristics of the anchor or merchant.

10. The device according to claim 9, wherein: According to the user's comment information and / or user behavior in the live broadcast room, the auxiliary live broadcast entrance is triggered to display: Detecting keywords in the comment information, and if the keywords match a preset question, determining a corresponding first entry copy and / or first entry image according to the preset question, and displaying an auxiliary live broadcast entry; and / or, Detect the user's silence time in the live broadcast room. If the silence time is greater than a preset time threshold, display the auxiliary live broadcast entrance with a preset second entrance copy and / or second entrance image.

11. The device according to claim 9, wherein: Determining a content display style according to the live broadcast room attribute, and displaying the recommended content to the user in the content display style includes: Using a voice conversion model corresponding to the attributes of the live broadcast room, and using a voice style corresponding to the anchor and / or the host, the recommended content is presented to the user in the form of voice playback; and / or, Using a text conversion model corresponding to the attributes of the live broadcast room, displaying the recommended content to the user in text form in a text style corresponding to the anchor and / or; and / or, The recommended content is displayed to the user in the form of images and / or videos using an image rendering model corresponding to the attributes of the live broadcast room and an image rendering style corresponding to the anchor and / or.

12. A method for processing auxiliary live broadcast, comprising: Obtain user input content and / or user behavior data during live broadcast, and identify user intent; Acquire recommended content based on the identified user intention information; wherein the user intention is obtained by performing user intention recognition on the user input content based on a semantic recognition model; The content display style is determined according to the attributes of the live broadcast room, and the recommended content is displayed to the user in the content display style. The content display style includes at least one of a voice style, a text style, and an image rendering style with personalized characteristics of the anchor or the merchant.

13. A processing device for auxiliary live broadcast, comprising: The intention recognition module is used to obtain user input content and / or user behavior data during the live broadcast and perform user intention recognition; A content recommendation module, used to obtain recommended content based on the identified user intention information; wherein the user intention is obtained by performing user intention recognition on the user input content according to the semantic recognition model; The content display module is used to determine the content display style according to the attributes of the live broadcast room, and display the recommended content to the user in the content display style. The content display style includes at least one of the voice style, text style and image rendering style with the personalized characteristics of the anchor or merchant.

14. A method for processing auxiliary live broadcast, comprising: In response to a user's query request, creating a conversation space between the virtual anchor and the user; According to the query request, obtaining recommended content by the method described in any one of claims 1 to 8; In the conversation space, the virtual anchor presents the recommended content to the user in a live broadcast through a one-to-one conversation mode, and continues to obtain and present the recommended content based on the user's comment information on the presented recommended content, and repeats the above actions until the user stops replying to information or a reply message to terminate the conversation is detected.

15. A method for processing auxiliary live broadcast, comprising: In response to a user's query request in the live broadcast room, entering a virtual display interface; According to the query request, obtaining recommended content by the method described in any one of claims 1 to 8; In the virtual display interface, the associated virtual models and / or virtual models related to the recommended content are displayed, triggering the virtual anchor to introduce the recommended content in combination with the virtual models and / or virtual models.

16. The method according to claim 15, wherein: Also includes: In response to a user's click operation on the virtual model and / or virtual mannequin, the virtual anchor is triggered to introduce the content of the part corresponding to the click operation.

17. An electronic device comprising: Memory, used to store programs; A processor, used to run the program stored in the memory to execute the auxiliary live broadcast processing method described in any one of claims 1 to 8, 12, and 14 to 16.

Images