Interaction method for video and processing method and device for video

Abstract

The present disclosure provides an interactive method for video and a processing method, device, equipment, medium, and program product for video, which relate to the field of artificial intelligence technology, especially to the technical fields of human-computer interaction, natural language processing, and large models, and can be used in application scenarios such as generative search, intelligent document editing, intelligent assistants, and intelligent e-commerce. The specific implementation scheme of the interactive method for video is: in response to the playback of the video, an interactive entry link is displayed in the playback page of the video; and in response to the selection operation of the interactive entry link, a first interactive page is displayed in a predetermined area of ​​the playback page except the playback area of ​​the video, wherein the interactive object associated with the first interactive page includes a virtual object corresponding to the character in the video; the virtual object is determined based on the video content of the video using a large language model.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to technical fields such as human-computer interaction, natural language processing and large models, and can be used in application scenarios such as generative search, intelligent document editing, intelligent assistants, and intelligent e-commerce. Background Art

[0002] With the development of computer technology and network technology, the application of artificial intelligence technology is becoming more and more extensive, and artificial intelligence technology has made breakthrough progress in various fields. Among them, artificial intelligence generated content (AI generated content, AIGC) is an important direction of deep learning.

[0003] With the rapid development of AIGC technology, for example, intelligent dialogue can be conducted based on AIGC technology to provide intelligent services. Summary of the invention

[0004] The present invention aims to provide a video interaction method and a video processing method, device, equipment, medium and program product that can improve the immersion and participation of watching videos, so as to solve the technical problem of low immersion and participation of users when watching videos due to unilateral interaction in the related art.

[0005] According to a first aspect of the present invention, there is provided an interactive method for a video, comprising: in response to the playback of a video, displaying an interactive entry link in a playback page of the video; and in response to a selection operation on the interactive entry link, displaying a first interactive page in a predetermined area of ​​the playback page other than a playback area of ​​the video, wherein the interactive objects associated with the first interactive page include virtual objects corresponding to characters in the video; the virtual objects are determined using a large language model based on the video content of the video.

[0006] According to a second aspect of the present invention, a method for processing a video is provided, comprising: using a large language model to determine virtual objects included in interactive objects associated with an interactive page based on the video content of the video; the virtual objects correspond to characters in the video; and in response to receiving a request to obtain the video, sending identification information of the video and the virtual objects to a terminal device that requests the video.

[0007] According to a third aspect of the present invention, there is provided an interactive device for videos, comprising: a first display module for displaying an interactive entry link in a playback page of the video in response to the playback of the video; and a second display module for displaying a first interactive page in a predetermined area of ​​the playback page other than the playback area of ​​the video in response to a selection operation on the interactive entry link, wherein the interactive objects associated with the first interactive page include virtual objects corresponding to characters in the video; the virtual objects are determined based on the video content of the video using a large language model.

[0008] According to a fourth aspect of the present invention, there is provided a processing device for a video, comprising: an object determination module for determining virtual objects included in interactive objects associated with an interactive page based on the video content of the video using a large language model; the virtual objects correspond to characters in the video; and an information sending module for sending identification information of the video and the virtual objects to a terminal device requesting the video in response to receiving a request to obtain the video.

[0009] According to a fifth aspect of the present invention, there is provided an electronic 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 so that the at least one processor can execute the interactive method for video or the processing method for video provided by the present invention.

[0010] According to a sixth aspect of the present invention, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to execute the interactive method for video or the processing method for video provided by the present invention.

[0011] According to a seventh aspect of the present invention, there is provided a computer program product, comprising a computer program / instruction, wherein the computer program / instruction is stored on at least one of a readable storage medium and an electronic device, and wherein the computer program / instruction, when executed by a processor, implements the interactive method for video or the processing method for video provided by the present invention.

[0012] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The accompanying drawings are used to better understand the present invention and do not constitute a limitation of the present invention.

[0014] Figure 1It is a schematic diagram of an application scenario of an interactive method for a video and a processing method and device for a video according to an embodiment of the present invention;

[0015] Figure 2 is a flow chart of a video interaction method according to an embodiment of the present invention;

[0016] Figure 3 is a flow chart of a method for processing a video according to an embodiment of the present invention;

[0017] Figure 4 is an interactive schematic diagram of an interactive method for video according to a first embodiment of the present invention;

[0018] Figure 5 is an interactive schematic diagram of an interactive method for video according to a second embodiment of the present invention;

[0019] Figure 6 is an interactive schematic diagram of an interactive method for video according to a third embodiment of the present invention;

[0020] Figure 7 is an interactive schematic diagram of an interactive method for video according to a fourth embodiment of the present invention;

[0021] Figure 8 is a schematic diagram of the principle of determining reply information according to an embodiment of the present invention;

[0022] Fig. 9 is a structural block diagram of a video interaction device according to an embodiment of the present invention;

[0023] Fig.10 is a structural block diagram of a video processing device according to an embodiment of the present invention; and

[0024] Fig.11 It is a block diagram of an electronic device for implementing the video interaction method or the video processing method according to the embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0026] The current interactive mode of video is single, and users can only passively accept content, lacking personalization and interactivity. The one-way interactive mode will not only reduce the user's immersion and participation, but may also cause users to lose interest in the video playback platform, thereby reducing the activity and user stickiness of the video playback platform.

[0027] For example, in the existing interactive methods, users can like their favorite videos to express their preference for the video content, but this interactive method is one-sided, and the user's immersion and participation are low. Users can also send real-time comments during video playback, but this interactive process requires users to provide real-time comment content, and users often give up due to the high discussion cost, resulting in low user immersion and participation.

[0028] In order to solve the problems existing in the related art, the present invention provides an interactive method for video and a processing method, device, equipment, medium and program product for video. Figure 1 The application scenarios of the method and device provided by the present invention are described.

[0029] Figure 1 The present invention is a schematic diagram of an application scenario of a video interaction method and a video processing method and device according to an embodiment of the present invention.

[0030] like Figure 1 As shown, the application scenario 100 of this embodiment may include a terminal device 110, which may be various electronic devices with display functions and interactive functions, including but not limited to smart phones, tablet computers, portable computers, desktop computers, servers, etc.

[0031] For example, a client application or search engine for playing videos may be installed in the terminal device 110, and the video may be any video such as an advertisement video, a TV series, a movie, or a short video, which is not limited in the present invention. For example, in an application scenario, a short video platform may be installed in the terminal device 110, and the video is a video played on the short video platform.

[0032] In one embodiment, the terminal device 110 can, for example, display an interactive entry link to the user 120 when playing a video to the user 120, and in response to the user 120 selecting the interactive entry link, display an interactive page in a predetermined area of ​​the video playback page other than the video playback area, so that the user can interact with the virtual object corresponding to the character in the video. The virtual object may include an object corresponding to a single character in the video, or may include multiple objects corresponding to multiple characters in the video, and the present invention is not limited to this.

[0033] like Figure 1As shown, the application scenario 100 may also include a server 130. The terminal device 110 may be connected to the server 130 through a network. For example, the server 130 may be a background management server for providing support for the operation of the client application installed in the terminal device. The server 130 may also be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server 130 may also be a server of a distributed system, or a server combined with a blockchain.

[0034] The video played by the terminal device 110 may be obtained by sending an acquisition request to the server 130. The server 130 may, for example, use a large language model to determine the reply information fed back by the virtual object to the user during the interaction between the user and the virtual object based on the video content of the video. When the virtual object includes multiple objects, the large language model may also determine which of the multiple objects feeds back the reply information to the user. The server 130 may also generate interaction guidance information based on the real-time played video content to guide the user to interact with the virtual object.

[0035] It should be noted that the interactive method for video provided by the present invention can be executed by the terminal device 110. Accordingly, the interactive device for video provided by the present invention can be set in the terminal device 110. The processing method for video provided by the present invention can be executed by the server 130. Accordingly, the processing device for video provided by the present invention can be set in the server 130.

[0036] It should be understood that Figure 1 The number and type of the terminal devices 110 and the server 130 in the embodiment are only for illustration. According to the implementation requirements, there may be any number and type of the terminal devices 110 and the server 130.

[0037] The following will be combined Figure 2 The interactive method for videos provided by the present invention is described in detail.

[0038] Figure 2 It is a flowchart of a video interaction method according to an embodiment of the present invention.

[0039] like Figure 2 As shown, the video interaction method 200 of this embodiment may include operations S210 to S220.

[0040] In operation S210, in response to the playing of the video, an interactive entry link is displayed in the playing page of the video.

[0041] In operation S220, in response to a selection operation on the interactive entry link, a first interactive page is displayed in a predetermined area of ​​the play page except for a play area of ​​the video.

[0042] According to an embodiment of the present invention, the interactive entrance may be, for example, a group chat entrance corresponding to the currently playing video, or a single chat entrance corresponding to the currently playing video. Accordingly, after detecting a selection operation on the interactive entrance link, the first interactive page added to the playback page may be a group chat page, or a single chat page. The selection operation on the interactive entrance link may be a single-click operation, a double-click operation, etc., which is not limited by the present invention.

[0043] For example, the interactive entry link may be displayed at any position in the video playback page, such as near the right edge of the playback page, or near the bottom of the playback page, etc., and the present invention is not limited to this.

[0044] In one embodiment, before displaying the first interactive page, the video playback area may be the entire playback page. When the first interactive page needs to be displayed, the terminal device may reduce the video playback area and display the first interactive page in any area of ​​the playback page except the video playback area. For example, when playing a video in portrait mode, the first interactive page may be displayed below the playback area; when playing a video in landscape mode, the first interactive page may be displayed on the left or right side of the playback area, and the present invention does not limit this.

[0045] According to an embodiment of the present invention, the interactive objects associated with the first interactive page may include virtual objects corresponding to the characters in the video in addition to the viewing objects of the video. There may be one or more virtual objects, and the virtual objects may be digital people generated in advance based on the character information of the characters in the video. Specifically, the virtual objects may be determined based on the video content of the video using a large language model, that is, the virtual objects are virtual characters constructed in advance by the large language model based on the video content. Thus, in the first interactive page, the dialogue information provided by the virtual objects is generated by the large language model for constructing the virtual characters.

[0046] In one embodiment, the base model can be trained using the role information and video content of the role in the video, so that the model learns the characteristics of the role in the video, thereby obtaining a large language model corresponding to the virtual object. The role information can include, for example, corpus data such as the lines of the role in the video, and can also include behavioral data such as the action and expression of the role, which is not limited in the present invention. Among them, the base model can be any open source pre-trained large language model, which is not limited in the present invention.

[0047] In order to facilitate the implementation of the interactive method for video, the present invention also provides a processing method for video. Figure 3 The video processing method provided by the present invention is described in detail.

[0048] Figure 3 It is a flowchart of a method for processing a video according to an embodiment of the present invention.

[0049] like Figure 3 As shown, the video processing method 300 of this embodiment may include operations S310 to S320.

[0050] In operation S310, a large language model is used to determine virtual objects included in an interactive object associated with an interactive page based on video content of a video.

[0051] In operation S320, in response to receiving a request for obtaining a video, the video and identification information of the virtual object are sent to a terminal device that requests the video.

[0052] According to an embodiment of the present invention, the server can construct one or more digital humans corresponding to the characters in each video in the resource library according to the video content of the video, and use the constructed digital humans as virtual objects. For example, the number of constructed digital humans can be input into the large language model as prompt information, because the large language model provides guidance information, so that the number of digital humans constructed by the large language model meets actual needs.

[0053] In one embodiment, the server can use the role information and video content of the role in the video to train the base model, so that the model learns the characteristics of the role in the video, thereby obtaining a large language model corresponding to the virtual object. The video content can include one or more of a video synopsis, a video script, and a plot summary.

[0054] According to actual needs, when the interactive page to be displayed is a single chat page, the number of virtual objects determined is one. When the interactive page to be displayed is a group chat page, the number of virtual objects determined is multiple, for example, 3, 4, 5, etc. Among them, the page type of the interactive page to be displayed can be determined according to the settings of the video viewing object on the video playback platform, and can also be dynamically adjusted according to the actual interactive traffic, and the present invention does not limit this.

[0055] After determining the virtual object for the video, the server can send the video and identification information of the virtual object constructed for the video to the terminal device in response to the acquisition request for the video sent by the terminal device, so as to display the interactive entry link when the terminal device plays the video.

[0056] According to an embodiment of the present invention, the identification information of a virtual object may be, for example, an image of the virtual object, or a role name of a key role corresponding to the virtual object, a still of the corresponding key role, etc., and the present invention is not limited thereto.

[0057] pass Figure 2 and Figure 3 According to the process described, an embodiment of the present invention can display an interactive page associated with a virtual object corresponding to a character in a video on a video playback page, and provide an interactive entry link to the interactive page when playing the video, so that the video viewing object can interact with the virtual object corresponding to the character in the video in real time, provide an intelligent interactive experience for the video viewing object, and help improve the video viewing object's participation and immersion in watching the video, thereby improving the attractiveness and user stickiness of the video playback platform.

[0058] The following will be combined Figure 4~Figure 7 The interaction principle of the interactive method for video provided by the present invention is further described and expanded.

[0059] Figure 4 4 is an interactive schematic diagram of an interactive method for video according to a first embodiment of the present invention.

[0060] like Figure 4 As shown, in an embodiment 400, the first interactive page displayed for the interactive method of the video may be a group chat page. Specifically, when playing a video, the playback page of the video may be page 410. In the page 410, the interactive entry link displayed is a link 411 of the group chat entry. In response to the selection operation of the link 411 of the group chat entry, the playback page of the video may be switched from page 410 to page 420. A first interactive page 421 is displayed in the page 420. The first interactive page 421 is a group chat page, and the group chat content is displayed in the group chat page. The interactive object associated with the first interactive page 421 may include one virtual object or multiple virtual objects. The interactive object associated with the first interactive page 421 may also include one or more video viewing objects in the video viewing object that participate in the interactive process by selecting the interactive entry link, and the present invention does not limit this.

[0061] In one embodiment, in addition to the group chat content, the group chat page may also include, for example, a preset group chat name displayed at the top of the page and / or identification information of an interactive object associated with the group chat, etc., which is not limited in the present invention.

[0062] In one embodiment, the link 411 of the group chat entrance may be displayed in text form or image form, etc. The displayed text or image may be pre-set. Alternatively, the identification information of the virtual object associated with the group chat may be used as the display content of the link 411 of the group chat entrance to attract the video viewing object to participate in the group chat.

[0063] In the case where there are multiple virtual objects associated with the group chat, for example, the identification information of the multiple objects can be alternately used as the display content of the link 411 of the group chat entrance to give the video viewing object the feeling that the group chat is in progress, further increasing the possibility of the viewing object participating in the group chat, and further improving the user experience. Alternatively, the object that has spoken the most recently among the multiple objects included in the virtual object can be determined based on the group chat content displayed in the first interactive page, and the object that has spoken the most recently can be used as the display content of the link 411 of the group chat entrance. It can be understood that the first interactive page here can be a group chat page that can be seen by other viewing objects except the viewing object of page 420.

[0064] Figure 5 4 is an interactive schematic diagram of an interactive method for video according to a second embodiment of the present invention.

[0065] like Figure 5 As shown, in one embodiment 500, the first interactive page displayed for the interactive method of the video can be a single chat page. Specifically, when playing a video, the playback page of the video can be page 510, in which the interactive entry link displayed is a link 511 of the single chat entry. In response to the selection operation of the link 511 of the single chat entry, the playback page of the video can be switched from page 510 to page 520. A first interactive page 521 is displayed in the page 520. The first interactive page 521 is a single chat page, and the single chat content is displayed in the single chat page. The interactive object associated with the first interactive page 521 includes a virtual object corresponding to the target character in the video and a viewing object of the video displayed on page 520. Among them, the target character is one of the multiple characters in the video.

[0066] In one embodiment, in addition to the single chat content, the single chat page may also include the name of the target character displayed at the top of the page and / or identification information of the virtual object corresponding to the target character, etc., which is not limited to the present invention.

[0067] In one embodiment, the link 511 of the single chat entrance can be displayed in text form or image form, etc. The displayed text or image can be pre-set. Alternatively, the identification information of the virtual object associated with the single chat can be used as the display content of the link 511 of the single chat entrance to attract the video viewing object to participate in the single chat.

[0068] In one embodiment, the target role can be determined by the large language model based on the video content and the portrait information of the viewing object of the video. Specifically, the server can determine the viewing object of the video through the acquisition request for the video sent by the terminal device, thereby obtaining the portrait information of the viewing object. Subsequently, the server can input the portrait information and video content of the viewing object into the large language model, and the large language model determines the virtual object associated with the single chat page. Among them, the portrait information can, for example, reflect the static attributes and dynamic attributes of the viewing object. The dynamic attributes can, for example, reflect the actors and the likes of the viewing object, and the present invention does not limit this. In this embodiment, the aforementioned operation S310 can, for example, be executed again in response to receiving the acquisition request to determine the virtual object in combination with the portrait information. For example, the large language model can use the virtual object of the role played by the actor liked by the viewing object in the video as the final determined virtual object. Through this embodiment, the possibility of the viewing object participating in the single chat can be increased, which is conducive to improving the user's participation and immersion in the playing video.

[0069] According to an embodiment of the present invention, after the viewing object of the video participates in a group chat or a single chat, the terminal device can detect that the viewing object inputs information for the first interactive page, and after determining that the input of the information is completed, the received input information is displayed as query information on the first interactive page. At the same time, the terminal device can send the query information to the server, and the server uses a large language model to process the received query information and generate reply information corresponding to the query information. Subsequently, the server can send the generated reply information to the terminal device. After receiving the reply information corresponding to the query information, the terminal device can display the reply information on the first interactive page.

[0070] For example, the server may input the query information sent by the terminal device into the large language model, and the large language model may output the reply information. In one embodiment, the terminal device may also send the context information of the query information (e.g., one or more pieces of information displayed before the query information and one or more pieces of information displayed after the query information) to the server. The server may input the query information and the context information into the large language model, and the large language model may generate the reply information based on the query information and the context information. By considering the context information of the query information when generating the reply information, the accuracy of the generated reply information may be improved.

[0071] It can be understood that when the first interactive page is a group chat page, the server can also synchronize the query information to all terminal devices displaying the group chat page after receiving the query information sent by the terminal device. After receiving the query information, each terminal device can display the query information on the group chat page, thereby achieving the effect of multiple viewers participating in the group chat.

[0072] In one embodiment, when the first interactive page is a group chat page, and the associated virtual objects include multiple objects corresponding to multiple roles in the video, the server can use the large language model to generate the reply information, and for example, can also use the large language model to determine the object associated with the reply information. For example, the server can input the query information and the context information into the large language model, and the large language model generates the reply information and the indication information of the virtual object corresponding to the reply information. The virtual object corresponding to the reply information is one of the multiple objects corresponding to the multiple roles mentioned above, for example, it can be the first object. For example, the input of the large language model can also include guidance information for guiding the generation of the reply information (for example, it can be a guide phrase "determine the role and reply content of the reply query information XXX in combination with the following context"). Subsequently, the server can send the identification information of the first object together with the reply information to the terminal device. In this way, after receiving the reply information and the identification information of the first object associated with the reply information, the terminal device can display the reply information and the identification information of the first object in association in the first interactive page. In this way, the effect can be achieved that the reply information in the first interactive page is provided by the characters in the video, giving the viewers the feeling of being able to chat with the characters in the video, thereby helping to increase the user's sense of immersion in interacting with the characters in the video during the video viewing process, and improving user activity and user stickiness of the video playback platform.

[0073] In one embodiment, when the server uses a large language model to generate reply information, for example, it can also consider the emotional information of the interactive object corresponding to the query information. That is, the emotional state of the viewing object that inputs the query information. The emotional information can be determined, for example, by the query information and the context of the query information. For example, it can be determined based on the interjections, expressions, etc. in all the information input by the interactive object corresponding to the query information. By considering the emotional information, the generated reply information can be made more likely to resonate with the viewing object, and the degree of interest of the video viewing object in chatting with the virtual object can be increased, thereby further improving the immersion and participation of the video viewing object in the video viewing process, and improving the user stickiness of the video playback platform.

[0074] For example, the server can input the query information and context information into a pre-built sentiment classification model, and the sentiment classification model outputs the sentiment category. Subsequently, the server can input the query information, context information and sentiment category into a large language model, and the large language model generates reply information that matches the sentiment category. Alternatively, the server can, for example, first input the query information and context information and corresponding guide information (for example, it can be a guide sentence "determine the sentiment category of the inquirer based on the following information") into a large language model, and the large language model outputs the sentiment information of the interactive object corresponding to the query information. The sentiment information can, for example, include sentiment categories and reasons for being classified into the sentiment category. Subsequently, the server can input the sentiment information, query information and context information into a large language model, and for example, can also input guide information for guiding the generation of reply information (for example, it can be a guide sentence "combining the following sentiment information and context, generate reply information for the query information XXX"), and the large language model outputs the reply information.

[0075] In one embodiment, when determining the reply information, the server may also consider the portrait information of the interactive object corresponding to the query information. For example, the server may input the query information, context information, and the portrait information of the interactive object corresponding to the query information into the large language model, and the large language model generates the reply information. In this way, the generated reply information is more likely to resonate with the viewing object. In some embodiments, if the query information involves items worn, worn, or used by the characters in the video, the reply information generated by the large language model often includes description information of these items. In this case, the server may use the information output by the large language model as the reply content, and when the reply content involves item information, retrieve the pre-built item library or online shopping platform based on the item information to retrieve the item link corresponding to the item information. Subsequently, the server may send both the item link and the reply content as reply information to the terminal device. In this way, the terminal device may display the reply content and the item link in association. In this way, the benefits of video conversion can be increased to a certain extent, and consumption conversion can be promoted. Considering that the generation of the reply content takes into account the portrait information of the interactive object corresponding to the query information, the items involved in the reply content are more likely to be items of interest to the video viewing object, which is conducive to promoting consumption conversion.

[0076] Figure 6 4 is an interactive schematic diagram of an interactive method for video according to a third embodiment of the present invention.

[0077] like Figure 6As shown, in an embodiment 600, when the first interactive page is a group chat page, in response to a selection operation on an interactive entry link, the play page can be switched to page 610. The first interactive page 611 displayed in page 610 can display continuously generated group chat content. The group chat content includes the conversation content of one or more virtual objects corresponding to the characters in the video. For example, it can include the conversation content of the second object among multiple virtual objects ((e.g., "XXXXXX"), the conversation content is displayed in association with the identification information of the virtual object), and the conversation content is displayed in association with the identification information 612 of the second object.

[0078] In this embodiment 600, the video viewing object can select the identification information 612 of the second object by an interactive method such as single-clicking or double-clicking to enter the single chat page with the second object. Specifically, the terminal device can switch the video playback page from page 610 to page 620 in response to the selection operation of the identification information 612 of the second object. A second interactive page 621 is displayed in the page 620, and the interactive object associated with the second interactive page 621 is the second object and the viewing object of the video. For example, if the character in the video corresponding to the second object is character A, then the second interactive page 621 can display the name of the character corresponding to the second object (such as character A) and / or the identification information of the second object in addition to the content of the conversation between the second object and the video viewing object. In this way, the video viewing object can switch the display page from the group chat interface to the single chat page with the character according to actual needs to meet the multiple needs of the video viewing object, further improve the immersion and participation of the video viewing object in watching the video, and improve the experience of the video viewing object in watching the video.

[0079] Figure 7 4 is an interactive schematic diagram of a video interactive method according to a fourth embodiment of the present invention.

[0080] like Figure 7 As shown, in an embodiment 700, the server and the terminal device can synchronize the progress of the terminal device playing the video and the interaction between the terminal device and the video viewing object. For example, when the video viewing object does not initiate participation in the chat with the virtual object corresponding to the character in the video, the server can generate interaction guidance information based on the video content played in real time by the terminal device to guide the video viewing object to participate in the group chat of the associated virtual object or initiate a single chat with the virtual object.

[0081] For example, the server can input the real-time video content into the large language model, and the large language model generates interaction guidance information. It is understandable that the information input into the large language model can also include guidance information (for example, it can be a guidance sentence "generate chat content to initiate a chat based on the following video content" or "generate a discussion / comment topic based on the following video content", etc.) to guide the large language model to generate interaction guidance information. After generating the interaction guidance information, the server can send the interaction guidance information to the terminal device for display by the terminal device. Through the display of the interaction guidance information, the video viewing object can be guided to participate in the group chat of the associated virtual object or initiate a single chat with the virtual object, so as to guide the video viewing object to improve the immersion and participation in the video viewing process.

[0082] In one embodiment, after the real-time video content is input into the large language model, the large language model can determine the virtual object corresponding to the interaction guidance information in addition to generating the interaction guidance information. For example, the determined virtual object can be the second object among the multiple virtual objects corresponding to the multiple characters in the video. For example, the virtual object corresponding to the interaction guidance information can be determined based on the indication information of the character or the indication information of the virtual object output by the large language model. Subsequently, the server can send the interaction guidance information and the identification information of the second object to the terminal device.

[0083] In the embodiment 700, when the video is played, the video playing page may be page 710, and in the page 710, the interactive entry link 711 displayed may be a link of a group chat entry or a link of a single chat entry. The following description is made by taking the interactive entry link 711 as a link of a group chat entry as an example.

[0084] After the terminal device receives the interactive guidance information sent by the server and the identification information of the virtual object associated with the interactive guidance information, the identification information of the virtual object can be used as the display content of the interactive entry link to display the interactive entry link. For example, if before receiving the interactive guidance information, the display content of the interactive entry link 711 is the identification information of the first object, and the received identification information is the identification information of the second object, then the play page can be switched from page 710 to page 720. In the page 720, the display content of the interactive entry link 721 is the identification information of the second object. The terminal device can also display the interactive guidance information 722 in the surrounding area of ​​the display area of ​​the interactive entry link 721 in page 720. In this way, the video viewing object can be provided with the feeling that the interactive guidance information 722 is initiated by the character in the video corresponding to the second object, that is, the feeling that the character in the video actively initiates a chat to the video viewing object can be provided, so that the probability of the user participating in the chat with the virtual object can be increased, which is conducive to improving the immersion and participation of the video viewing object when watching the video.

[0085] In this embodiment 700, if the video viewing object performs a selection operation such as single-clicking or double-clicking on the interactive entry link 721 in the page 720, the play page can be switched from the page 720 to the page 730. The page 730 displays a first interactive page 731, and the terminal device can also display the interactive guidance information and the second corresponding identification information in association in the first interactive page 731. This further provides the video viewing object with the feeling that the character in the video actively initiates a chat with him.

[0086] The following will be combined Figure 8 The principle of determining the reply information is further described and expanded.

[0087] Figure 8 It is a schematic diagram of the principle of determining reply information according to an embodiment of the present invention.

[0088] In one embodiment, multimodal reply information can also be provided for the interaction process. For example, the reply information provided can include information of one or more modes among multiple modes such as text mode, video mode, audio mode, image mode, etc. In this way, the chat content and form between the video viewing object and the virtual object are enriched, so as to improve the realism of the user's chat with the character in the video, and improve the user's immersion and participation.

[0089] For example, when the server uses a large language model to determine the reply information, it can first input the query information and context information into the large language model, and the large language model generates text information. For example, guidance information can also be input into the large language model to guide the large language model to decide the modality of the final reply information. The guidance information may include, for example, "based on the following context, determine the modality of the generated reply information; if it is a text modality, output the reply information; if it is a non-text modality, output the call information required to generate the reply information of the non-text modality and the type of the non-text modality", etc., and the present invention does not limit this. After the large language model generates text information, the server can, for example, determine whether the call information is included based on the text information. If the text information includes the modality type, it can be determined that the text information includes the call information, and the other content in the text information except the modality type is used as the call information, and the information acquisition tool corresponding to the modality type is called, and the information acquisition tool obtains the reply information of the non-text modality.

[0090] In one embodiment, the reply information may include both text modal information and non-text modal information. The server may use the information in the text information except the modal type and call information, and the non-text modal information obtained by the information acquisition tool as the reply information.

[0091] For example, if the non-text mode is an audio mode, the text information output by the large language model may include text content corresponding to the audio type reply information. The server may use the text content as the input of a pre-set audio generation tool, and the audio generation tool converts the text content into audio. Among them, the audio generation tool may be, for example, any tool or deep learning model built based on text-to-speech (TTS) technology, and the present invention is not limited to this. When generating audio, for example, the timbre information of the character in the video corresponding to the virtual object corresponding to the reply information may also be considered. Specifically, the server may extract the timbre features of multiple characters in the video based on the audio in the video in advance. The text information output by the large language model may also include indication information of the character in the video corresponding to the reply information. The server may obtain the timbre features of the corresponding character based on the indication information, and use the obtained timbre features and text content as the input of the audio generation tool, so that the generated audio has the timbre features of the corresponding character, thereby further improving the fidelity of the generated reply information and improving the immersion of the video viewing object.

[0092] For example, if the non-text modality is a video modality, the text information output by the large language model may include text content corresponding to the video type reply information. The server may search the video library based on the text content, and use the retrieved video as the reply content of the video modality. If the video is not retrieved, the server may use the text content as input to a pre-set video generation tool, and the video generation tool may convert the text content into a video. The video generation tool may be, for example, any existing tool that can convert text into a video, and the present invention is not limited to this.

[0093] For example, if the non-text modality is an image modality, the text information output by the large language model may include image description information. The server may search the image library based on the image description information, and use the retrieved image as the reply content of the image modality. In the case where the image is not retrieved, the server may use the image description information as the input of a pre-set image generation tool, and the image generation tool generates an image based on the image description information. The image generation tool may, for example, be any existing tool that can generate images based on text, and the present invention does not limit this. It is understandable that the reply information of the image modality may, for example, be an image in any format, such as a moving picture or a still picture, and the reply information of the image modality may also be an image in an emoticon package, etc., and the present invention does not limit this.

[0094] like Figure 8 As shown, in an embodiment 800 , the server may be integrated with an end-to-end framework, which may include a large language model 810 and a tool component 820 .

[0095] The input of the large language model 810 may include input information 801 such as the aforementioned query information and context information. The output of the large language model 810 is text information 802. According to actual needs, the tools in the tool component 820 can be called based on the text information. For example, the image retrieval tool in the tool component can be called to retrieve the reply information of the image modality, and the image generation tool in the tool component can be called to generate the reply information of the image modality. The video retrieval tool in the tool component can be called to retrieve the reply information of the video modality, and the video generation tool in the tool component can be called to generate the reply information of the video modality. The audio generation tool in the tool component can also be called to generate the reply information of the audio modality. In this process, the large language model 810 can act as an agent.

[0096] In one embodiment, the end-to-end framework may further include a central control component 830. When generating text information 802, the large language model 810 may, for example, act as a proxy for the search engine 831 in the central control component 830 to search for knowledge from a database 832 and generate text information 802 based on the searched knowledge. Whether it is necessary to search for knowledge from the database 832 is determined by the large language model 810 through decision making.

[0097] It should be noted that the guidance information mentioned in the above embodiments is only used as an example to facilitate understanding of the present invention, and the present invention does not limit this. In the embodiments where the guidance information is not mentioned, if the large language model is used to determine or generate relevant content, the guidance information that guides the large language model to generate relevant content can be input into the large language model according to actual needs, which will not be described in detail here.

[0098] It is understandable that the aforementioned video may be an advertising video. Through the aforementioned embodiments, the effect of real interaction between the characters in the advertising video and the advertising viewers can be achieved, thereby improving the advertising conversion rate to a certain extent.

[0099] Based on the interactive method for video provided by the present invention, the present invention also provides an interactive device for video. Fig. 9 The device is described in detail.

[0100] Fig. 9 4 is a structural block diagram of a video interaction device according to an embodiment of the present invention.

[0101] like Fig. 9 As shown, the interactive device 900 for video in this embodiment may include a first presentation module 910 and a second presentation module 920 .

[0102] The first display module 910 is used to display the interactive entry link in the video playback page in response to the video playback. In one embodiment, the first display module 910 can be used to perform the operation S210 described above, which will not be repeated here.

[0103] The second display module 920 is used to display the first interactive page in a predetermined area other than the video playback area in the playback page in response to the selection operation of the interactive entry link. The interactive objects associated with the first interactive page include virtual objects corresponding to the characters in the video; the virtual objects are determined based on the video content of the video using a large language model. In one embodiment, the second display module 920 can be used to perform the operation S220 described above, which will not be repeated here.

[0104] According to an embodiment of the present invention, the virtual object includes multiple objects corresponding to multiple characters in the video. The first display module 910 is specifically used to alternately use the identification information of the multiple objects as the display content of the interactive entry link to display the interactive entry link.

[0105] According to an embodiment of the present invention, the second display module 920 is also used to display the query information in the first interactive page in response to receiving the query information input for the first interactive page. The interactive device 900 for video may also include an information sending module for sending the query information and the context information of the query information. Accordingly, the second display module 920 may also be used to display the reply information in the first interactive page in response to receiving the reply information corresponding to the query information. The reply information is determined based on the query information and the context information using a large language model.

[0106] According to an embodiment of the present invention, the virtual object includes multiple objects corresponding to multiple characters in the video. The second display module 920 can be specifically used to: in response to receiving the identification information of the first object associated with the reply information, display the reply information and the identification information of the first object in a related manner in the first interactive page. The first object is one of the multiple objects; the first object is determined by using a large language model based on the query information and the context information.

[0107] According to an embodiment of the present invention, the reply information includes at least one of the following types of information: text type, audio type, video type, and image type. Among them, the information of other types than text type included in the reply information is obtained by calling the corresponding information acquisition tool by using the calling information; the calling information is determined by using a large language model based on the query information and context information.

[0108] According to an embodiment of the present invention, the above-mentioned interactive device 900 for video may also include a third display mode, which is used to display a second interactive page associated with the second object and the viewing object of the video in a superimposed manner in the playback page in response to a selection operation on the identification information of the second object displayed in the first interactive page. The second object is one of the multiple objects.

[0109] According to an embodiment of the present invention, the first display module 910 can also be used to: in response to receiving the interaction guide information and the identification information of the third object associated with the interaction guide information, display the interaction entry link using the identification information of the third object as the display content of the interaction entry link; and display the interaction guide information in the surrounding area of ​​the display area of ​​the interaction entry link. The interaction guide information and the third object are determined based on the real-time video content using a large language model; and the third object is one of the virtual objects associated with the first interaction page.

[0110] According to an embodiment of the present invention, the second display module 920 may also be used to display the interaction guidance information and the identification information of the third object in a related manner in the first interaction page in response to the display of the first interaction page when the interaction guidance information is displayed in the surrounding area.

[0111] According to an embodiment of the present invention, the reply information includes reply content and item links. The reply content is generated by using a large language model based on query information, context information and the portrait information of the video viewing object; the item link is retrieved based on the reply content.

[0112] According to an embodiment of the present invention, the above-mentioned reply information is determined by the large language model based on the query information, the context information and the emotion information of the interactive object corresponding to the query information. The emotion information is determined by the large language model based on the query information and the context information.

[0113] According to an embodiment of the present invention, the virtual object corresponds to a target character in a video; the target character is one of multiple characters in the video; the target character is determined by a large language model based on the video content and the portrait information of the viewing object of the video.

[0114] Based on the video processing method provided by the present invention, the present invention also provides a video processing device. Fig.10 The device is described in detail.

[0115] Fig.10 4 is a structural block diagram of a video processing device according to an embodiment of the present invention.

[0116] like Fig.10As shown, the video processing device 1000 of this embodiment may include an object determination module 1010 and an information sending module 1020 .

[0117] The object determination module 1010 is used to determine the virtual objects included in the interactive objects associated with the interactive page based on the video content of the video using the large language model. The virtual objects correspond to the characters in the video. In one embodiment, the object determination module 1010 can be used to perform the operation S310 described above, which will not be repeated here.

[0118] The information sending module 1020 is used to send the video and the identification information of the virtual object to the terminal device requesting the video in response to receiving the acquisition request for the video. In one embodiment, the information sending module 1020 can be used to perform the operation S320 described above, which will not be repeated here.

[0119] According to an embodiment of the present invention, the video processing device 1000 may further include an information determination module, which is used to determine reply information based on the query information and the context information in response to receiving the query information and the context information sent by the terminal device. The information sending module 1020 may also be used to send the reply information to the terminal device.

[0120] According to an embodiment of the present invention, the information determination module may also be used to determine the first object associated with the reply information based on the query information and context information using a large language model when the query information is information input for an interactive page associated with multiple objects. The information sending module 1020 may also be used to send the identification information of the first object to the terminal device in association with the reply information. The virtual object includes multiple objects corresponding to multiple characters in the video; the first object is one of the multiple objects.

[0121] According to an embodiment of the present invention, the information determination module may include: an information generation submodule for generating text information based on query information and context information using a large language model; an information acquisition submodule for, when the text information includes call information, using the call information to call a corresponding information acquisition tool to obtain information of other types other than text types included in the reply information; and an information determination submodule for, when the text information does not include the call information, using the text information as the reply information. The other types include at least one of the following types: audio type, video type, image type.

[0122] According to an embodiment of the present invention, the information determination module may include: an information generation submodule, which is used to generate reply content based on query information, context information, and portrait information of the video viewing object using a large language model; and a link acquisition submodule, which is used to retrieve an item link corresponding to the item information based on the item information in response to the reply content involving item information. The reply information includes the reply content and the item link.

[0123] According to an embodiment of the present invention, the above-mentioned information determination module may include: an emotion determination sub-module, which is used to use a large language model to determine the emotion information of the interactive object corresponding to the query information based on the query information and context information; and an information generation sub-module, which is used to use a large language model to determine the reply information based on the query information, context information and emotion information.

[0124] According to an embodiment of the present invention, the information determination module may also be used to: determine the interaction guidance information and the second object corresponding to the interaction guidance information based on the video content played in real time by the terminal device using a large language model. The information sending module may also be used to send the interaction guidance information and the identification information of the second object to the terminal device. The second object is one of the virtual objects.

[0125] It should be noted that the collection, storage, use, processing, transmission, provision, disclosure and application of user personal information involved in the technical solution of the present invention are in compliance with the provisions of relevant laws and regulations, necessary confidentiality measures are taken, and they do not violate public order and good morals. In the technical solution of the present invention, the user's authorization or consent is obtained before obtaining or collecting user personal information.

[0126] According to an embodiment of the present invention, the present invention also provides an electronic device, a readable storage medium and a computer program product.

[0127] Fig.11 A schematic block diagram of an example electronic device 1100 that can be used to implement an interactive method for a video or a processing method for a video of an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0128] like Fig.11As shown, the device 1100 includes a computing unit 1101, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1102 or a computer program loaded from a storage unit 1108 to a random access memory (RAM) 1103. In the RAM 1103, various programs and data required for the operation of the device 1100 can also be stored. The computing unit 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0129] A number of components in the device 1100 are connected to the I / O interface 1105, including: an input unit 1106, such as a keyboard, a mouse, etc.; an output unit 1107, such as various types of displays, speakers, etc.; a storage unit 1108, such as a disk, an optical disk, etc.; and a communication unit 1109, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1109 allows the device 1100 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0130] The computing unit 1101 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1101 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1101 performs the various methods and processes described above, such as an interactive method for a video or a processing method for a video. For example, in some embodiments, an interactive method for a video or a processing method for a video may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 1108. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 1100 via the ROM 1102 and / or the communication unit 1109. When the computer program is loaded into the RAM 1103 and executed by the computing unit 1101, one or more steps of the interactive method for a video or the processing method for a video described above may be performed. Alternatively, in other embodiments, the computing unit 1101 may be configured to execute an interactive method for a video or a processing method for a video in any other appropriate manner (for example, by means of firmware).

[0131] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0132] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0133] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0134] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0135] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0136] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. Among them, the server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server may also be a server of a distributed system, or a server combined with a blockchain.

[0137] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0138] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An interactive method for a video, characterized in that: The method comprises: In response to the playing of the video, displaying the interactive entry link in the playing page of the video, including: using identification information of a plurality of objects alternately as display content of the interactive entry link, and displaying the interactive entry link; and In response to a selection operation of the interactive entry link, displaying a first interactive page in a predetermined area of ​​the play page except for a play area of ​​the video; In response to a selection operation on identification information of a second object displayed in the first interactive page, a second interactive page associated with the second object and a viewing object of the video is displayed in a superimposed manner in the play page; Among them, the first interaction page is a group chat page, and the interaction objects associated with the first interaction page include virtual objects corresponding to the characters in the video; the virtual objects are determined based on the video content of the video using a large language model, and the virtual objects include the multiple objects corresponding one-to-one to the multiple characters in the video; the second interaction page is a single chat page with the second object; the second object is one of the multiple objects.

2. The method according to claim 1, characterized in that The method further comprises: In response to receiving query information input for the first interactive page, displaying the query information in the first interactive page; Sending the query information and context information of the query information; and In response to receiving reply information corresponding to the inquiry information, displaying the reply information in the first interactive page; The reply information is determined by using the large language model based on the query information and the context information.

3. The method according to claim 2, characterized in that The virtual objects include a plurality of objects corresponding to the plurality of characters in the video; and displaying the reply information in the first interactive page includes: In response to receiving identification information of a first object associated with the reply information, displaying the reply information and the identification information of the first object in association with each other in the first interactive page, The first object is one of the multiple objects; the first object is determined by using the large language model based on the query information and the context information.

4. The method according to claim 2, characterized in that: The reply information includes at least one of the following types of information: text type, audio type, video type, image type; Among them, the reply information includes other types of information except the text type, which is obtained by calling the corresponding information acquisition tool using the calling information; the calling information is determined by using the large language model based on the query information and the context information.

5. The method according to claim 1, characterized in that The method further comprises: In response to receiving the interaction guide information and the identification information of the third object associated with the interaction guide information, displaying the interaction entry link using the identification information of the third object as the display content of the interaction entry link; and The interactive guidance information is displayed in the peripheral area of ​​the display area of ​​the interactive entry link, Among them, the interaction guidance information and the third object are determined by using the large language model based on the real-time played video content; and the third object is one of the virtual objects associated with the first interaction page.

6. The method according to claim 5, characterized in that The method further comprises: In the case where the interaction guide information is displayed in the peripheral area, in response to the display of the first interaction page, the interaction guide information and the identification information of the third object are displayed in association in the first interaction page.

7. The method according to claim 2, characterized in that: The reply information includes reply content and item link; The reply content is generated by using the large language model based on the query information, the context information and the portrait information of the viewing object of the video; the item link is retrieved based on the reply content.

8. The method according to claim 2, characterized in that: The reply information is determined by the large language model based on the query information, the context information and the emotion information of the interaction object corresponding to the query information; The sentiment information is determined by the large language model based on the query information and the context information.

9. The method according to claim 1, characterized in that: The virtual object corresponds to a target character in the video; the target character is one of a plurality of characters in the video; The target role is determined by the large language model based on the video content and the portrait information of the viewing object of the video.

10. A method for processing a video, characterized in that: The method comprises: Using a large language model based on the video content of the video, determine the virtual object included in the interactive object associated with the interactive page; the virtual object corresponds to the character in the video; In response to receiving a request for obtaining the video, sending the video and identification information of the virtual object to a terminal device that requests the video; In response to receiving the query information sent by the terminal device and the context information of the query information, using the large language model to determine reply information based on the query information and the context information, and sending the reply information to the terminal device; In the case where the inquiry information is information input for an interactive page associated with a plurality of objects, the large language model is used to determine a first object associated with the reply information based on the inquiry information and the context information, and identification information of the first object is sent to the terminal device in association with the reply information; wherein the virtual object includes the plurality of objects corresponding one-to-one to the plurality of characters in the video, and the first object is one of the plurality of objects; The large language model is used to determine interaction guidance information and a second object corresponding to the interaction guidance information based on the video content played in real time by the terminal device, and the interaction guidance information and identification information of the second object are sent to the terminal device to guide the video viewing object to participate in a group chat associated with the virtual object or initiate a single chat with the second object; wherein the second object is one of the virtual objects.

11. The method according to claim 10, characterized in that Using the large language model to determine the response information based on the query information and the context information includes: Generate text information based on the query information and the context information using the large language model; In the case where the text information includes calling information, using the calling information to call a corresponding information acquisition tool to acquire other types of information included in the reply information except the text type; In the case where the text message does not include the calling information, using the text message as the reply information, The other types include at least one of the following types: audio type, video type, and image type.

12. The method according to claim 10, characterized in that Using the large language model to determine the response information based on the query information and the context information includes: Using the large language model to generate a response content based on the query information, the context information, and the portrait information of the viewing object of the video; In response to the reply content involving item information, obtaining an item link corresponding to the item information based on the item information, The reply information includes the reply content and the item link.

13. The method according to claim 10, characterized in that Using the large language model to determine the response information based on the query information and the context information includes: Determining the emotion information of the interactive object corresponding to the query information based on the query information and the context information using the large language model; and The large language model is used to determine the response information based on the query information, the context information, and the sentiment information.

14. An interactive device for video, characterized in that: The device comprises: A first display module is used to display an interactive entry link in a playback page of the video in response to the playback of the video, including: using identification information of multiple objects alternately as display content of the interactive entry link to display the interactive entry link; and A second display module, configured to display the first interactive page in a predetermined area of ​​the play page except for the play area of ​​the video in response to a selection operation on the interactive entry link; A third display module, configured to display a second interactive page associated with the second object and a viewing object of the video in a superimposed manner on the playback page in response to a selection operation on the identification information of the second object displayed in the first interactive page; Among them, the first interaction page is a group chat page, and the interaction objects associated with the first interaction page include virtual objects corresponding to the characters in the video; the virtual objects are determined based on the video content of the video using a large language model, and the virtual objects include the multiple objects corresponding one-to-one to the multiple characters in the video; the second interaction page is a single chat page with the second object; the second object is one of the multiple objects.

15. A video processing device, characterized in that: The device comprises: An object determination module, configured to determine a virtual object included in an interactive object associated with an interactive page based on the video content of the video using a large language model; the virtual object corresponds to a character in the video; An information sending module, configured to send the video and identification information of the virtual object to a terminal device that requests the video in response to receiving a request for obtaining the video; A first module is configured to, in response to receiving the query information and the context information of the query information sent by the terminal device, determine the reply information based on the query information and the context information using the large language model, and send the reply information to the terminal device; A second module is configured to determine, when the query information is information input for an interactive page associated with a plurality of objects, using the large language model to determine a first object associated with the reply information based on the query information and the context information, and to send identification information of the first object in association with the reply information to the terminal device; wherein the virtual object includes the plurality of objects corresponding one-to-one to the plurality of characters in the video, and the first object is one of the plurality of objects; The third module is used to use the large language model to determine the interaction guidance information and the second object corresponding to the interaction guidance information based on the video content played in real time by the terminal device, and send the interaction guidance information and the identification information of the second object to the terminal device to guide the video viewing object to participate in the group chat associated with the virtual object or initiate a single chat with the second object; wherein the second object is one of the virtual objects.

16. An electronic device, comprising: at least one processor; as well as 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 so that the at least one processor can execute any one of claims 1 to 13.

17. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 13.

18. A computer program product, comprising a computer program / instruction, wherein the computer program / instruction is stored on at least one of a readable storage medium and an electronic device, and when the computer program / instruction is executed by a processor, the steps of the method according to any one of claims 1 to 13 are implemented.

Images