Method and device for generating review information based on large model, electronic equipment and storage medium

Abstract

The present disclosure provides a large model-based comment information generation method and device, electronic equipment and storage medium, which are related to the technical field of artificial intelligence, and specifically related to the technical fields of deep learning, large model, natural language processing and the like. The specific scheme is: based on a large model, the to-be-commented resource is understood, and the description information of the to-be-commented resource is obtained; according to the description information, the comment information of the to-be-commented resource is obtained, the comment information at least includes the comment video of the to-be-commented resource, and the comment video is displayed in the comment area. The intelligent generation of comment video and text is realized, the accuracy of comment information can be improved, the comment generation operation is simplified, and the comment speed is improved. Moreover, by introducing the video comment form, more diverse comment forms are provided for users, and the user experience is greatly improved.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, particularly to the fields of deep learning, large models, and natural language processing, and specifically to a method, apparatus, electronic device, and storage medium for generating comment information based on a large model. Background Technology

[0002] In news feed applications, user comments are an important part of content consumption. In existing technologies, user comments are in text form, resulting in a relatively simple content format in the comment section. This often leads to poor interactivity of the application and fails to effectively increase the time users spend on the application. Summary of the Invention

[0003] This disclosure provides a method, apparatus, electronic device, and storage medium for generating comment information based on a large model.

[0004] According to one aspect of this disclosure, a method for generating comment information based on a large model is provided, comprising:

[0005] Based on a large model, we can understand the resources to be commented on and obtain descriptive information about them.

[0006] Based on the description information, obtain the comment information of the resource to be commented, wherein the comment information includes at least the comment video of the resource to be commented;

[0007] The video of the comment is displayed in the comments section.

[0008] According to another aspect of this disclosure, a comment information generation apparatus based on a large model is provided, comprising:

[0009] The first acquisition module is used to understand the resource to be commented based on the large model and obtain the descriptive information of the resource to be commented.

[0010] The second acquisition module is used to acquire comment information of the resource to be commented based on the description information, wherein the comment information includes at least the comment video of the resource to be commented.

[0011] The comment display module is used to display the comment videos in the comment section.

[0012] According to another aspect of this disclosure, an electronic device is provided, 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, the instructions being executed by the at least one processor to enable the at least one processor to perform the large model-based comment information generation method described in one aspect of the above-described embodiment.

[0013] According to another aspect of this disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are stored thereon for causing the computer to execute the large model-based comment information generation method described in the above-described embodiment.

[0014] According to another aspect of this disclosure, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the large model-based comment information generation method described in the above-described embodiment.

[0015] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0016] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0017] Figure 1 A flowchart illustrating a method for generating comment information based on a large model, provided in an embodiment of this disclosure;

[0018] Figure 2 A flowchart illustrating another method for generating comment information based on a large model, provided in this embodiment of the disclosure;

[0019] Figure 2a This is a schematic diagram illustrating a commentary video provided in an embodiment of this disclosure.

[0020] Figure 3 A flowchart illustrating another method for generating comment information based on a large model, provided in this embodiment of the disclosure;

[0021] Figure 4 A flowchart illustrating another method for generating comment information based on a large model, provided in this embodiment of the disclosure;

[0022] Figure 5 A flowchart illustrating another method for generating comment information based on a large model, provided in this embodiment of the disclosure;

[0023] Figure 6 This is a schematic diagram of the structure of a comment information generation device based on a large model, provided in an embodiment of this disclosure;

[0024] Figure 7 This is a block diagram of an electronic device used to implement the large model-based comment information generation method of the present disclosure embodiments. Detailed Implementation

[0025] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0026] The following description, with reference to the accompanying drawings, outlines a method, apparatus, and electronic device for generating comment information based on a large model, according to embodiments of the present disclosure.

[0027] Artificial intelligence (AI) is the study of enabling computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). It involves both hardware and software technologies. AI hardware technologies generally include computer vision, speech recognition, natural language processing, and related technologies such as deep learning, big data processing, and knowledge graphs.

[0028] Natural Language Processing (NLP) is an important field within computer science and artificial intelligence. It studies the theories and methods for enabling effective communication between humans and computers using natural language. NLP is a science that integrates linguistics, computer science, and mathematics. Its main applications include machine translation, public opinion monitoring, automatic summarization, opinion extraction, text classification, question answering, text semantic comparison, and speech recognition.

[0029] Deep learning (DL) is a new research direction in the field of machine learning (ML). It was introduced into machine learning to bring it closer to its original goal—artificial intelligence. Deep learning learns the inherent laws and hierarchical representations of sample data. The information gained during this learning process greatly aids in the interpretation of data such as text, images, and sound. Its ultimate goal is to enable machines to possess analytical and learning capabilities like humans, capable of recognizing data such as text, images, and sound. Deep learning is a complex machine learning algorithm that has achieved results in speech and image recognition far exceeding previous related technologies.

[0030] Large models refer to machine learning models with a massive number of parameters and complex computational structures. These models are typically built from deep neural networks and have billions or even hundreds of billions of parameters. The purpose of large models is to improve their expressive power and predictive performance, enabling them to handle more complex tasks and data. Large models have wide applications in various fields, including natural language processing, computer vision, speech recognition, and recommender systems. By training on massive amounts of data to learn complex patterns and features, large models possess stronger generalization capabilities and can make accurate predictions on unseen data.

[0031] Figure 1 This is a flowchart illustrating a method for generating comment information based on a large model, as provided in an embodiment of this disclosure.

[0032] like Figure 1 As shown, this method for generating comment information based on a large model may include, but is not limited to, the following steps:

[0033] S101, based on the large model, understand the resource to be commented and obtain the descriptive information of the resource to be commented.

[0034] It should be noted that the execution entity of the comment information generation method in this embodiment can be a hardware device with data processing capabilities and / or the necessary software to drive the hardware device. Optionally, the execution entity may include a server, computer, user terminal, and other smart devices. Optionally, the user terminal includes, but is not limited to, mobile phones, computers, smart voice interaction devices, etc. Optionally, the server includes, but is not limited to, a network server, an application server, or a server of a distributed system, or a server combined with blockchain, etc.

[0035] In some embodiments, the resource to be commented on can be a resource on various resource platforms, including but not limited to: news clients, social media platforms, video sharing websites, etc. The resource to be commented on can include, but is not limited to: articles, images, audio, and video. For example, an article can be text materials from an academic forum. Another example is an image resource published on a social media platform. Yet another example is audio, which can be songs, music, radio dramas, etc. And yet another example is video, which can be short videos, or resources such as TV series and movies.

[0036] In some embodiments, the resource to be commented on can be understood as a resource that a user is browsing on a certain platform, such as a short video currently playing on a social media platform.

[0037] In some embodiments, the descriptive information of the resource to be commented on may include, but is not limited to, key information and topic information of the resource to be commented on. It is understood that the key information and topic information of the resource to be commented on can serve as the basis for selecting subsequent comment videos and generating comment text.

[0038] In some embodiments, key information may include, but is not limited to, keywords, title, and subject of the resource to be commented on.

[0039] In some embodiments, semantic analysis and content understanding can be performed on the resource to be commented based on a pre-trained large model to extract key information and topic tags from the resource to be commented.

[0040] S102, Based on the description information, obtain the comment information of the resource to be commented on, which includes at least the comment video of the resource to be commented on.

[0041] S103 displays comment videos in the comments section.

[0042] In some embodiments, comment information for the resource to be commented on can be output based on descriptive information and combined with a pre-trained large model. In this embodiment of the disclosure, the comment information includes at least the comment video of the resource to be commented on.

[0043] In some embodiments, the comment information for the resource to be commented on may also include comment text; that is, the comment information for the resource to be commented on may be a combination of comment video and comment text.

[0044] In some embodiments, comment videos for the resource to be commented on are obtained from a video library based on the description information. Optionally, the resource to be commented on is episode i of a TV series, and the comment video can be episode i+1 or episode -1 of the same TV series. Optionally, the comment video can include, but is not limited to, other viewpoint videos on the same event, videos extending the background of the event, etc.

[0045] In some embodiments, after obtaining the video to be commented on, a first comment text for the resource to be commented on can be generated based on the resource to be commented on and the video to be commented on. It is understood that the first comment text is a transitional comment text, which can be used to guide users from the resource to be commented on to the video to be commented on, improving the coherence and readability of the comment.

[0046] In some embodiments, a second comment text for the resource to be commented on can be generated based on the description information, and a comment video for the resource to be commented on can be generated based on the second comment text.

[0047] In some embodiments, the comment section corresponding to a user may include at least two types of comment display areas, such as, but not limited to, text display areas and video display areas.

[0048] In some embodiments, after obtaining the comment information of the resource to be commented on, each type of comment content in the comment information can be filled into the corresponding display area.

[0049] The comment information generation method according to embodiments of this disclosure achieves intelligent generation of comment videos and text, improving the accuracy and intelligence level of comment information. Furthermore, by introducing video comment formats, it provides users with more diverse comment options, greatly enhancing the user experience. Moreover, through intelligent comment information generation based on a large model, it simplifies the comment generation process, reduces comment complexity, not only increasing comment speed but also increasing the number of comments for resources, thereby making it easier to push resources to be commented on.

[0050] Figure 2 This is a flowchart illustrating a method for generating comment information based on a large model, as provided in an embodiment of this disclosure.

[0051] like Figure 2 As shown, this method for generating comment information based on a large model may include, but is not limited to, the following steps:

[0052] S201, Obtain description information for the resource to be commented on.

[0053] For optional implementations of step S201, please refer to [link / reference]. Figure 1 Optional implementation methods of step S101, and Figure 1 Other related parts in the embodiments involved will not be described in detail here.

[0054] S202, Based on the description information, retrieve the comment video of the resource to be commented on from the video library.

[0055] In some embodiments, candidate videos related to the resource to be commented are obtained from a video library based on the description information. Further, the video with the highest relevance is selected from the videos as the comment video for the resource to be commented.

[0056] Optionally, the description information of each video in the video library is obtained, and a relevance calculation is performed based on the description information of the resource to be evaluated and the description information of the videos to obtain the relevance between the resource to be evaluated and each video. Further, the videos in the database are filtered based on the video relevance to obtain candidate videos related to the resource to be evaluated. Optionally, videos with a relevance greater than or equal to a set value are selected as candidate videos.

[0057] Optionally, candidate videos can be sorted from highest to lowest, and the top-ranked candidate videos can be selected as comment videos for the videos to be commented on. This ensures that the comment videos best match the resources to be evaluated, making the evaluations of the resources more accurate and improving the user experience.

[0058] In this embodiment of the disclosure, the comment videos obtained from the video library may include the next episode of a movie or TV series, other viewpoint videos on the same event, videos extending the background of the event, etc. This not only enhances the interactivity and attractiveness of the comments, but also realizes the presentation of multimodal comments. Through the multimodal comment format, multi-dimensional comments on the resources to be commented on can be realized, making the evaluation of the resources to be commented on more comprehensive and accurate.

[0059] S203, Based on the resource to be commented on and the video to be commented on, generate the first comment text for the resource to be commented on.

[0060] In some embodiments, after acquiring the comment video, speech analysis and content understanding are performed on the resource to be commented on and the comment video based on a large language model, and a first comment text is generated based on the understood content. It is understood that the first comment text is a transitional comment text, which can be used to guide users from the resource to be commented on to the comment video, improving the coherence and readability of the comment. Optionally, the first comment text may also include a brief evaluation or opinion statement on the resource to be commented on and the comment video, realizing further commentary on the comment video, thereby allowing users to better understand the content of the resource to be commented on through the commentary on the comment video.

[0061] In some embodiments, historical comment data of the user is obtained, which may include, but is not limited to, historical resources that have been commented on, and the corresponding comment content. Further, the generated first comment text is polished based on the historical comment data and a public knowledge base. That is, by combining information such as the user's historical comment data and a public knowledge base, the first comment text is polished so that its expression logic, sentences, and grammar not only conform to human expression patterns but also align with the user's commenting habits, thereby improving the quality of the comment and making it more accurate.

[0062] In some embodiments, the expression logic, sentences, and syntax of the first comment text can be analyzed using a public knowledge base. If anomalies are found in these aspects of the first comment text, adjustments and optimizations are made to obtain an optimized comment text, ensuring that its expression logic, sentences, and syntax conform to human expression patterns. Furthermore, user comment preference information can be analyzed based on historical comment data, including but not limited to comment style and comment length. Based on this comment preference information, the style and length of the optimized comment text are further optimized to obtain the final first comment text.

[0063] S204, Displays the comment video and the first comment text in the comment section.

[0064] In some embodiments, the comment section for a user may include a text display area and a video display area.

[0065] In some embodiments, after obtaining the comment information of the resource to be commented on, each type of comment content can be filled into the corresponding display area. That is, the comment video is filled into the video display area, and the first comment text is filled into the text display area.

[0066] In some embodiments, the comment video can be played in response to another user swiping to the comment video.

[0067] In some embodiments, other users can click the play button on the comment video to trigger playback of the comment video, such as... Figure 2a As shown.

[0068] It is understood that in this embodiment of the application, the comment area on the front end not only has a text display area, but also adds a video display area, thereby supporting the display and playback of comment videos and ensuring that multimodal video comments are displayed normally to users.

[0069] In this embodiment, compared to users manually searching or entering comment content, a large-scale model is used to deeply understand the content of the resource to be evaluated and select comment videos matching the resource from a massive video library. This means it can automatically recommend comment videos related to the resource to be evaluated to the user. Simultaneously, combining the resource to be evaluated and the comment videos, high-quality transitional comment text is generated to improve the coherence and readability of the comments. This embodiment achieves intelligent generation of comment videos and comment text, improving the accuracy and intelligence of comment information. Furthermore, by introducing video comment formats, more diverse comment formats are provided for users to choose from, greatly improving the user experience. Moreover, intelligent comment information generation based on a large-scale model simplifies the comment generation operation, reduces comment complexity, not only increasing comment speed but also increasing the number of comments for the resource, thus making it easier for the resource to be evaluated to be pushed out.

[0070] Figure 3 This is a flowchart illustrating a method for generating comment information based on a large model, as provided in an embodiment of this disclosure.

[0071] like Figure 3 As shown, this method for generating comment information based on a large model may include, but is not limited to, the following steps:

[0072] S301, Obtain the description information of the resource to be commented on.

[0073] For optional implementations of step S301, please refer to [link / reference]. Figure 1 Optional implementation methods of step S101, and Figure 1 Other related parts in the embodiments involved will not be described in detail here.

[0074] S302, Based on the description information, generate the second comment text for the resource to be commented on.

[0075] In some embodiments, historical comment data of users is obtained, which may include, but is not limited to, previously commented resources and their corresponding comment content. Further, a second comment text is generated based on descriptive information, historical comment data, and a public knowledge base. That is, a deep understanding of the resource to be commented on, combined with information such as user historical comment data and a public knowledge base, is used to generate a second comment text related to the resource to be commented on using a large model. This ensures that the expression logic, sentences, and syntax of the second comment text not only conform to human expression patterns but also align with users' commenting habits, thereby improving the quality and accuracy of the comments.

[0076] In some embodiments, an initial third comment text is generated based on the description information. Further, the initial third comment text is refined based on historical comment data and a public knowledge base to generate a second comment text. Optionally, the initial third comment text can first be analyzed using a public knowledge base to examine its expression logic, sentences, syntax, etc. If anomalies exist in these aspects of the initial third comment text, adjustments and optimizations are made to obtain an optimized fourth comment text, ensuring that the expression logic, sentences, and syntax of the fourth comment text conform to human expression patterns. Further, user comment preference information can be analyzed based on historical comment data, such as, but not limited to, comment style and comment length. Based on this comment preference information, the optimized fourth comment text is further optimized in style and length to obtain the final second comment text.

[0077] Optionally, the second comment text can be a statement of opinion on the evaluated resource, an expression of emotion, or an extended discussion, which can further express the user's emotions.

[0078] S303, Based on the second comment text, generate a comment video for the resource to be commented on.

[0079] In some embodiments, the second comment text can be input into a pre-trained text-to-video generation model to obtain a comment video of the resource to be commented on. Generating comment videos from the second comment text allows the acquisition of comment videos to be independent of existing video resources, enriching the content of the comment videos and reducing the probability of duplicate comment videos for the same resource.

[0080] In some embodiments, the second comment text can be input into a pre-trained text-to-video generation model to obtain an initial first video. Further, to enhance the display effect and quality of the comment video, at least one of background audio and special effects can be added to the first video to obtain a second video. The second video can then undergo image quality enhancement to obtain the comment video for the resource to be commented on. That is, only background audio can be added to the first video, or only features of the first video can be added, or both background audio and features of the first video can be added simultaneously. Optionally, the background audio can be background music, voice-over, or narration, etc.

[0081] In some embodiments, a target background audio and a target special effect are selected from candidate background audio and candidate special effects based on at least one content in the first video and the second comment text, and added to the first video to obtain the second video. That is, the target background audio and target special effect are selected from candidate background audio and candidate special effects based on the understanding of at least one content in the first video and the second comment text using a large model.

[0082] In some embodiments, a user's selection operation is received, a target background audio and a target special effect are determined based on the selection operation, and added to a first video to obtain a second video.

[0083] In some embodiments, historical or current image enhancement operations are obtained, and target image quality parameters corresponding to the first video are determined based on the historical or current image enhancement operations. Further, image enhancement is performed on the second video based on the target image quality parameters to obtain the comment video of the resource to be commented on.

[0084] In some embodiments, the preferred image quality can be determined based on the user's historical image quality enhancement operations. The most frequently used image quality parameters from the historical operations can be selected as the target image quality parameters, making the image quality more in line with the user's habits. This not only improves the quality of the video being reviewed but also enhances the user experience.

[0085] In some embodiments, initial image quality parameters of a first video are obtained, and in response to the initial image quality parameters not meeting the image quality requirements, image quality enhancement is performed on a second video to obtain a comment video for the resource to be commented on.

[0086] In this embodiment of the disclosure, by adding background audio and features to the comment video, the elements of the comment video become more comprehensive, and the image quality can be enhanced. This not only increases the quality of the comment video but also makes the comment video more intelligent.

[0087] S304, Display the comment video and the second comment text in the comment section.

[0088] For optional implementations of step S304, please refer to [link / reference]. Figure 2 Optional implementation methods of step S204, and Figure 1 Other related parts in the embodiments involved will not be described in detail here.

[0089] In this embodiment, compared to users manually searching or entering comment content, a large model is used to deeply understand the content of the resource to be evaluated, generating high-quality comment text. Then, a comment video matching the resource to be evaluated is generated based on the comment text. In this embodiment, generating video from text makes the generation of comment videos more efficient. Especially on social media platforms, the number of comments is often related to the popularity of a resource. Intelligent commenting can simplify the commenting process, increase commenting speed, and quickly increase the number of comments, thus making it easier for the resource to be commented on to be pushed. Furthermore, the comment video can be optimized to further improve its quality, thereby enhancing user satisfaction and experience. Moreover, by introducing video commenting, users are provided with more diverse commenting options, further improving the user's commenting experience.

[0090] Understandably, users can freely choose between the two comment video generation methods mentioned above. A selection option could be provided on the front-end interface, allowing users to choose their preferred generation method from the two options, increasing user freedom and improving the user experience. Alternatively, if a suitable comment video cannot be selected from the video library, a second comment text could be used to generate the comment video. Generating video from text not only produces high-quality comment videos but also simplifies and simplifies the generation process, improving the intelligence level of the comment video generation.

[0091] Based on the above embodiments, comment information of the resource to be commented on can be stored, such as... Figure 4 As shown, the process of storing comment information for a resource to be commented on may include, but is not limited to, the following steps:

[0092] S401, Based on the load balancing strategy, determine the target storage node for the comment information from the storage nodes in the distributed storage system.

[0093] In some embodiments, the backend server can store the comment videos in a distributed storage system, which may include multiple storage nodes. A load balancing strategy can be used to determine a suitable target storage node from among these nodes to store the comment information of the resource to be commented on. For example, load balancing can be performed based on the remaining storage space of the storage nodes to obtain the target storage node. The distributed storage system ensures the security of the comment information of the resource to be commented on, avoiding loss due to contention for storage space.

[0094] S402, Obtain the current network status information, and determine the conversion information of the comment information based on the network status information.

[0095] S403: Send the comment information of the resource to be commented to the target storage node.

[0096] In some embodiments, current network status information is obtained, and the conversion information of the comment information is determined based on the network status information. Optionally, the network status information may include, but is not limited to, network bandwidth, traffic volume, network bandwidth utilization, etc. Optionally, the conversion information of the comment information may include, but is not limited to, format conversion information and compression type, etc.

[0097] Furthermore, the comment information is transformed and compressed based on the transformation information to obtain the target comment information, and then the target comment information is sent to the target storage node.

[0098] It is understandable that the comment information of the resource to be commented on can be read from the target storage node during subsequent browsing and restored to obtain the original comment information of the resource to be commented on.

[0099] In this embodiment of the disclosure, distributed storage can ensure the security of comment information storage. Furthermore, by considering network and device conditions, appropriate formats can be selected for conversion and compression, saving resource consumption during uploading and further guaranteeing the security of comment information during the uploading process.

[0100] Figure 5 This is a flowchart illustrating a method for generating comment information based on a large model, as provided in an embodiment of this disclosure.

[0101] like Figure 5 As shown, this method for generating comment information based on a large model may include, but is not limited to, the following steps:

[0102] S501, retrieve the description information of the resource to be commented on.

[0103] For optional implementations of step S501, please refer to [link / reference]. Figure 1 Optional implementation methods of step S101, and Figure 1 Other related parts in the embodiments involved will not be described in detail here.

[0104] After obtaining the description information, you can execute steps S502-504 of branch 1 to determine the comment information of the resource to be commented on, or you can execute steps 505-509 of branch 2 to determine the comment information of the resource to be commented on.

[0105] S502, Based on the description information, retrieve candidate videos related to the resource to be commented on from the video library.

[0106] S503: Select the video with the highest relevance from the candidate videos as the comment video for the resource to be commented on.

[0107] S504, Based on the resource to be commented on and the video to be commented on, generate the first comment text for the resource to be commented on.

[0108] For optional implementations of steps S502 to S504, please refer to [link / reference]. Figure 2 Optional implementation methods of steps S202-203, and Figure 2 Other related parts in the embodiments involved will not be described in detail here.

[0109] S505, Based on the description information, generate the initial third comment text.

[0110] S506, based on historical comment data and a public knowledge base, polishes the third comment text to generate the second comment text.

[0111] S507, generate the initial first video based on the second comment text.

[0112] S508 adds background audio and effects to the first video to obtain the second video.

[0113] S509, enhance the image quality of the second video to obtain the comment video of the resource to be commented on.

[0114] For optional implementations of steps S505 to S509, please refer to [link / reference]. Figure 3 Optional implementation methods of steps S302-303, and Figure 3 Other related parts in the embodiments involved will not be described in detail here.

[0115] S510 displays comment information for resources in the comment section.

[0116] For optional implementations of step S510, please refer to [link / reference]. Figure 2 Optional implementation methods of step S204, and Figure 2 Other related parts in the embodiments involved will not be described in detail here.

[0117] S511, Based on the load balancing strategy, determine the target storage node for the comment information from the storage nodes in the distributed storage system.

[0118] S512: Obtain the current network status information, and based on the network status information, determine the conversion and compression of the comment information.

[0119] S513 sends comment information for the resource to be commented to the target storage node.

[0120] In this embodiment, the determination of comment videos and the generation of text are realized, which improves the accuracy and intelligence of comment information. Furthermore, by introducing video comment formats, users are provided with more diverse comment options, greatly enhancing the user experience. Moreover, intelligent comment information generation based on a large model simplifies the comment generation process, reduces comment complexity, and not only increases comment speed but also increases the number of comments for resources, making it easier to push resources to be commented on.

[0121] Corresponding to the comment information generation methods based on large models provided in the above embodiments, an embodiment of this disclosure also provides a comment information generation device based on large models. Since the information processing device provided in this disclosure corresponds to the comment information generation methods based on large models provided in the above embodiments, the implementation methods of the above-mentioned comment information generation methods based on large models are also applicable to the information processing device provided in this disclosure, and will not be described in detail in the following embodiments.

[0122] Figure 6 This is a schematic diagram of a comment information generation device based on a large model, provided in an embodiment of this disclosure.

[0123] like Figure 6 As shown, the comment information generation device 600 based on a large model according to this embodiment of the present disclosure includes a first acquisition module 601, a second acquisition module 602 and a comment display module 603.

[0124] The first acquisition module 601 is used to acquire the description information of the resource to be commented on;

[0125] The second acquisition module 602 is used to acquire the comment information of the resource to be commented based on the description information, wherein the comment information includes at least the comment video of the resource to be commented.

[0126] The comment display module 603 is used to display the comment video in the comment section.

[0127] In some embodiments, the second acquisition module 602 is further configured to acquire the comment video of the resource to be commented on from the video library based on the description information.

[0128] In some embodiments, the second acquisition module 602 is further configured to acquire candidate videos related to the resource to be commented from the video library according to the description information; and select the video with the highest relevance from the candidate videos as the comment video for the resource to be commented.

[0129] In some embodiments, the second acquisition module 602 is further configured to, after acquiring the comment video from the video library according to the description information, generate a first comment text for the resource to be commented based on the resource to be commented and the comment video.

[0130] In some embodiments, the second acquisition module 602 is further configured to generate a second comment text for the resource to be commented based on the description information; and generate a comment video for the resource to be commented based on the second comment text.

[0131] In some embodiments, the second acquisition module 602 is further configured to acquire the user's historical comment data and public knowledge base; and to polish the first comment text or the second comment text based on the historical comment data and public knowledge base.

[0132] In some embodiments, the second acquisition module 602 is further configured to generate an initial third comment text based on the description information; and to refine the third comment text based on the historical comment data and the public knowledge base to generate the second comment text.

[0133] In some embodiments, the second acquisition module 602 is further configured to input the second comment text into a pre-trained text-to-video generation model to obtain an initial first video; add at least one of background audio and special effects to the first video to obtain a second video; and enhance the image quality of the second video to obtain a comment video of the resource to be commented on.

[0134] In some embodiments, the second acquisition module 602 is further configured to select a target background audio from candidate background audio based on at least one content from the first video and the second comment text, and / or select a target effect from candidate effects and add it to the first video to obtain the second video.

[0135] In some embodiments, the second acquisition module 602 is further configured to receive a selection operation, determine the target background audio and target special effects based on the selection operation, and add them to the first video to obtain the second video.

[0136] In some embodiments, the second acquisition module 602 is further configured to acquire historical or current operations of image quality enhancement; determine target image quality parameters corresponding to the first video based on the historical or current operations of image quality enhancement; and perform image quality enhancement on the second video based on the target image quality parameters to obtain the comment video of the resource to be commented on.

[0137] In some embodiments, the second acquisition module 602 is further configured to acquire the initial image quality parameters of the first video, and in response to the initial image quality parameters not meeting the image quality requirements, to enhance the image quality of the second video to obtain the comment video of the resource to be commented on.

[0138] In some embodiments, the comment display module 603 is further configured to fill the video display area of ​​the comment section with the comment video.

[0139] In some embodiments, the second acquisition module 602 is further configured to determine the target storage node of the comment information from the storage nodes in the distributed storage system based on a load balancing strategy, and send the comment information of the resource to be commented to the target storage node.

[0140] In some embodiments, the second acquisition module 602 is further configured to acquire network status information, and determine the target format and compression information of the comment video in the comment information based on the network status information; convert and compress the comment video based on the target format and compression information to obtain target comment information, and send the target comment information to the target storage node.

[0141] The comment information generation apparatus according to embodiments of this disclosure realizes the determination of comment videos and the generation of text, thereby improving the accuracy and intelligence level of comment information. Furthermore, by introducing video comment formats, it provides users with more diverse comment formats for easier selection, greatly improving the user experience. Moreover, through intelligent comment information generation based on a large model, the comment generation operation is simplified, and the complexity of comments is reduced. This not only increases the speed of commenting but also increases the number of comments for resources, making it easier to push resources to be commented on.

[0142] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0143] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0144] Figure 7A schematic block diagram of an example electronic device 700 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0145] like Figure 7 As shown, device 700 includes a computing unit 701, which can perform various appropriate actions and processes based on computer programs / instructions stored in read-only memory (ROM) 702 or loaded from storage unit 706 into random access memory (RAM) 703. RAM 703 may also store various programs and data required for the operation of device 700. The computing unit 701, ROM 702, and RAM 703 are interconnected via bus 704. Input / output (I / O) interface 705 is also connected to bus 704.

[0146] Multiple components in device 700 are connected to I / O interface 705, including: input units 706 such as keyboard, mouse, etc.; output units 707 such as various types of displays, speakers, etc.; storage units 708 such as disks, optical disks, etc.; and communication units 709 such as network cards, modems, wireless transceivers, etc. Communication unit 709 allows device 700 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0147] The computing unit 701 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the various methods and processes described above, such as the large model-based comment information generation method. For example, in some embodiments, the large model-based comment information generation method can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 706. In some embodiments, part or all of the computer program / instructions can be loaded and / or installed on device 700 via ROM 702 and / or communication unit 709. When the computer program / instructions are loaded into RAM 703 and executed by the computing unit 701, one or more steps of the large model-based comment information generation method described above can be performed. Alternatively, in other embodiments, the computing unit 701 may be configured, by any other suitable means (e.g., by means of firmware), to perform a comment information generation method based on a large model.

[0148] Various embodiments 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-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: implementations in one or more computer programs / instructions that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transferring data and instructions to the storage system, the at least one input device, and the at least one output device.

[0149] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0150] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0151] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, 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 sound input, voice input, or tactile input).

[0152] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet, and blockchain networks.

[0153] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. The client-server relationship is created by computer programs / instructions running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0154] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in the disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this document does not impose any restrictions.

[0155] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for generating comment information based on a large model, wherein, The method includes: Based on a large model, we can understand the resources to be commented on and obtain descriptive information about them. Based on the description information, obtain the comment information of the resource to be commented, wherein the comment information includes at least the comment video of the resource to be commented; The video of the comment is displayed in the comment section; The step of obtaining the comment information of the resource to be commented based on the description information includes: Based on the description information, retrieve the comment video for the resource to be commented on from the video library; After retrieving the comment video from the video library based on the description information, the process further includes: Based on the resource to be commented on and the video to be commented on, a first comment text is generated for the resource to be commented on, and the first comment text is used to guide the user from the resource to be commented on to the video to be commented on.

2. The method according to claim 1, wherein, The step of retrieving the comment video of the resource to be commented from the video library based on the description information includes: Based on the description information, candidate videos related to the resource to be commented are obtained from the video library; The video with the highest relevance is selected from the candidate videos and used as the comment video for the resource to be commented on.

3. The method according to claim 1, wherein, The step of obtaining the comment information of the resource to be commented based on the description information includes: Based on the description information, generate a second comment text for the resource to be commented on; Based on the second comment text, a comment video for the resource to be commented on is generated.

4. The method according to claim 1 or 3, wherein, The method further includes: Access users' historical comment data and public knowledge base; Based on the historical comment data and public knowledge base, the first or second comment text is polished.

5. The method according to claim 4, wherein, The process of generating the second comment text includes: Based on the described information, generate the initial third comment text; Based on the historical comment data and public knowledge base, the third comment text is polished to generate the second comment text.

6. The method according to claim 3, wherein, The step of generating a comment video for the resource to be commented on based on the second comment text includes: The second comment text is input into the pre-trained text-to-video generation model to obtain the initial first video; Add at least one of background audio and special effects to the first video to obtain the second video; The second video is enhanced to obtain the comment video for the resource to be commented on.

7. The method according to claim 6, wherein, Adding at least one of background audio and special effects to the first video to obtain the second video includes: Based on at least one content from the first video and the second comment text, a target background audio is selected from candidate background audio, and / or a target effect is selected from candidate effects, and added to the first video to obtain the second video.

8. The method according to claim 6, wherein, Adding at least one of background audio and special effects to the first video to obtain the second video includes: The system receives a selection operation, determines the target background audio and target special effects based on the selection operation, and adds them to the first video to obtain the second video.

9. The method according to claim 6, wherein, The step of enhancing the image quality of the second video to obtain the comment video for the resource to be commented on includes: Retrieve historical or current image enhancement operations; Based on the historical or current operations of the image enhancement, the target image quality parameters corresponding to the first video are determined; The second video is enhanced based on the target image quality parameters to obtain the comment video of the resource to be commented on.

10. The method according to claim 6 or 9, wherein, The step of enhancing the image quality of the second video to obtain the comment video for the resource to be commented on includes: The initial image quality parameters of the first video are obtained. In response to the initial image quality parameters not meeting the image quality requirements, the image quality of the second video is enhanced to obtain the comment video of the resource to be commented on.

11. The method according to any one of claims 1-3 or 6-9, wherein, The display of the comment video in the comment section includes: The comment video is then added to the video display area of ​​the comment section.

12. The method according to any one of claims 1-3 or 6-9, wherein, The method further includes: Based on a load balancing strategy, the target storage node for the comment information is determined from the storage nodes in the distributed storage system, and the comment information for the resource to be commented is sent to the target storage node.

13. The method according to claim 12, wherein, Sending the comment information of the resource to be commented to the target storage node includes: Obtain network status information, and based on the network status information, determine the target format and compression information of the comment video in the comment information; The comment video is converted and compressed based on the target format and compression information to obtain target comment information, and then the target comment information is sent to the target storage node.

14. A comment information generation device based on a large model, wherein, The device includes: The first acquisition module is used to understand the resource to be commented based on the large model and obtain the descriptive information of the resource to be commented. The second acquisition module is used to acquire comment information of the resource to be commented based on the description information, wherein the comment information includes at least the comment video of the resource to be commented. The comment display module is used to display the comment videos in the comment section; The second acquisition module is further configured to: Based on the description information, retrieve the comment video of the resource to be commented on from the video library; The second acquisition module is further configured to: Based on the resource to be commented on and the video to be commented on, a first comment text is generated for the resource to be commented on, and the first comment text is used to guide the user from the resource to be commented on to the video to be commented on.

15. The apparatus according to claim 14, wherein, The second acquisition module is further configured to: Based on the description information, candidate videos related to the resource to be commented are obtained from the video library; The video with the highest relevance is selected from the candidate videos and used as the comment video for the resource to be commented on.

16. The apparatus according to claim 14, wherein, The second acquisition module is further configured to: Based on the description information, generate a second comment text for the resource to be commented on; Based on the second comment text, a comment video for the resource to be commented on is generated.

17. The apparatus according to claim 14 or 16, wherein, The second acquisition module is further configured to: Access users' historical comment data and public knowledge base; Based on the historical comment data and public knowledge base, the first or second comment text is polished.

18. The apparatus according to claim 17, wherein, The second acquisition module is further configured to: Based on the described information, generate the initial third comment text; Based on the historical comment data and public knowledge base, the third comment text is polished to generate the second comment text.

19. The apparatus according to claim 16, wherein, The second acquisition module is further configured to: The second comment text is input into the pre-trained text-to-video generation model to obtain the initial first video; Add at least one of background audio and special effects to the first video to obtain the second video; The second video is enhanced to obtain the comment video for the resource to be commented on.

20. The apparatus according to claim 19, wherein, The second acquisition module is further configured to: Based on at least one content from the first video and the second comment text, a target background audio is selected from candidate background audio, and / or a target effect is selected from candidate effects, and added to the first video to obtain the second video.

21. The apparatus according to claim 19, wherein, The second acquisition module is further configured to: The system receives a selection operation, determines the target background audio and target special effects based on the selection operation, and adds them to the first video to obtain the second video.

22. The apparatus according to claim 19, wherein, The second acquisition module is further configured to: Retrieve historical or current image enhancement operations; Based on the historical or current operations of the image enhancement, the target image quality parameters corresponding to the first video are determined; The second video is enhanced based on the target image quality parameters to obtain the comment video of the resource to be commented on.

23. The apparatus according to claim 19 or 22, wherein, The second acquisition module is further configured to: The initial image quality parameters of the first video are obtained. In response to the initial image quality parameters not meeting the image quality requirements, the image quality of the second video is enhanced to obtain the comment video of the resource to be commented on.

24. The apparatus according to any one of claims 14-16 or 19-22, wherein, The comment display module is also used for: The comment video is then added to the video display area of ​​the comment section.

25. The apparatus according to any one of claims 14-16 or 19-22, wherein, The second acquisition module is further configured to: Based on a load balancing strategy, the target storage node for the comment information is determined from the storage nodes in the distributed storage system, and the comment information for the resource to be commented is sent to the target storage node.

26. The apparatus according to claim 25, wherein, The second acquisition module is further configured to: Obtain network status information, and based on the network status information, determine the target format and compression information of the comment video in the comment information; The comment video is converted and compressed based on the target format and compression information to obtain target comment information, and then the target comment information is sent to the target storage node.

27. 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 to enable the at least one processor to perform the method as described in any one of claims 1-13.

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

29. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method steps of any one of claims 1-13.

Images