Method and system for converting patent text into video based on AI technology

Abstract

The invention provides a method and a system for converting a patent text into a video based on an AI technology. The method comprises the following steps: acquiring patent technology information and information used for generating copywriting keywords; generating picture information according to the patent technology information, generating copywriting information and audio information according to the information used for generating copywriting keywords, and synthesizing the picture information, the copywriting information and the audio information to obtain video information; and according to the generated picture information, copywriting information, audio information and video information, adjusting patent technology information and / or information used for generating copywriting keywords, and repeating the previous step until video conversion of the patent technology is completed. According to the patent technology information provided by the user, the keywords in the patent technology information can be extracted to automatically generate professional and accurate video content, the video production period is greatly shortened, meanwhile, the user can visually know the technology content, conversion of scientific and technological achievements is facilitated, and the patent technology is better promoted and popularized.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence and video technology, and in particular, to a method and system for converting patent text (patent documents and / or scientific and technological achievements) into video based on AI, and also to a corresponding computer conversion device and computer-readable storage medium. Background Art

[0002] As an innovative technological achievement protected by law, patent technology has multiple functions and significances, such as encouraging innovation, protecting innovative achievements, promoting technological exchanges and cooperation, enhancing corporate competitiveness, and promoting economic development. With the rapid development of science and technology, people's demand for patent technology is increasing day by day. However, traditional forms of patent information often have problems such as high reading threshold and single information presentation, which to a certain extent restricts the dissemination and application of patent technology. How to better promote and popularize patent technology has become an urgent problem to be solved in this field.

[0003] With the rapid development of artificial intelligence (AI) technology, especially the continuous advancement of natural language processing (NLP) and computer vision (CV) technology, it has become possible to convert text content into video form, which can not only enrich the way of information presentation, but also improve the efficiency of information dissemination and the audience's acceptance, thereby better meeting the needs of patent technology promotion and popularization.

[0004] However, although AI technology has shown great potential in video generation, how to ensure that the generated video content accurately reflects the technical points and ensures a high correlation between the generated video and the patent text has become a technical difficulty in the process of generating videos from patent texts. So far, no description or report of similar technology to the present invention has been found, and no similar materials at home and abroad have been collected. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method and system for converting patent text into video based on AI technology, and also provides a corresponding computer conversion device and computer-readable storage medium to solve the problems of high threshold for reading patent information and single information presentation method in the prior art.

[0006] According to one aspect of the present invention, a method for converting patent text into video based on AI technology is provided, comprising:

[0007] Obtain patent technology information and information used to generate copywriting keywords;

[0008] Generate image information according to the patent technology information, generate text information and audio information according to the information for generating text keywords, and synthesize the image information, text information and audio information to obtain video information;

[0009] According to the generated image information, text information, audio information and video information, adjust the patent technology information and / or the information used to generate text keywords, and repeat the previous step until the video conversion of the patent technology is completed.

[0010] Preferably, the patent technology information includes: patent technology text data and / or patent technology image data.

[0011] Preferably, the information used to generate copy keywords includes: keyword information and / or patent technology information set by the user.

[0012] Preferably, generating image information according to the patent technology information includes:

[0013] Preprocess the patented technical information to obtain key graphic information and key technical text information containing semantic structure;

[0014] Using the key graphic information to construct a visual element library;

[0015] Providing a patent technology information keyword library and a natural language processing model, using the patent technology information keyword library to train the natural language processing model to obtain a keyword extraction model;

[0016] Using the key technical text information as input to the keyword extraction model, and using the keyword extraction model's understanding of semantics to obtain technical keywords of the key technical text information;

[0017] According to the result of semantic understanding, key graphic and text information matching the technical keywords is selected from the visual element library to form a sample to be generated;

[0018] An image generation model is provided, and the sample to be generated is used as the input of the image generation model to obtain image information that is highly relevant to the patent technology information.

[0019] Preferably, the preprocessing of the patent technology information includes:

[0020] Obtain patent technology information in text format;

[0021] Extracting key graphic information from the patent technical information in text format for building a visual element library;

[0022] The patent technical information in the text format is preprocessed to obtain key technical text information and semantic structure of the patent technical information in the text format.

[0023] Preferably, generating copywriting information according to the information for generating copywriting keywords includes:

[0024] Providing a patent technology information keyword library and a natural language processing model, and using the patent technology information keyword library to train the natural language processing model to obtain a keyword extraction model;

[0025] The keyword extraction model is used to understand the semantics of the information used to generate the copywriting keywords, identify the copywriting keywords in the information used to generate the copywriting keywords, and then generate the copywriting information.

[0026] Preferably, the copywriting keywords include elements related to themes, concepts and / or relationships; wherein:

[0027] The subject corresponds to the patent name and technical effect in the patent technical information, the concept corresponds to the technical indicators in the patent technical information, and the relationship corresponds to the technical field and / or application scenario in the patent technical information.

[0028] Preferably, the audio information is generated according to the information for generating text keywords, and the audio information is generated based on the text information using text-to-speech technology.

[0029] Preferably, the synthesizing the picture information, text information and audio information to obtain the video information includes:

[0030] Computer vision technology is used to arrange the picture information, text information and audio information, add animation effects, and match sound effects and background music, ultimately generating patented technology video information with technical points and visual effects.

[0031] According to another aspect of the present invention, a system for converting patent text into video based on AI technology is provided, comprising:

[0032] A front-end interface unit, which is used to input patent technology information and information for generating copy keywords, and receive and generate image information, copy information, audio information and video information, and / or display video information;

[0033] A back-end server unit, which is used to generate image information according to the patent technology information, generate text information and audio information according to the information used to generate text keywords, synthesize the image information, text information and audio information to obtain video information, and feed it back to the front-end interface unit;

[0034] A database unit is used to store the patent technology information and information for generating copy keywords as well as the picture information, copy information, audio information and video information.

[0035] Preferably, the front-end interface unit includes:

[0036] A user upload module, which is used to input patent technology information and upload it to the back-end server unit, and at the same time output a request to generate image information of the patent technology information to the back-end server unit;

[0037] A copywriting generation module, the module is used to input information for generating copywriting keywords and upload it to the back-end server unit, and at the same time output a request to generate copywriting information, audio information and video information for the information for generating copywriting keywords to the back-end server unit;

[0038] A feedback and interaction module, which is used to provide an interactive interface and use the received image information, text information, audio information and video information to adjust the input patent technology information and / or information used to generate text keywords;

[0039] The display module is used to display the received video information.

[0040] Preferably, the backend server unit comprises:

[0041] An image generation module, the module is used to receive the patent technology information and a request for generating image information, and generate image information based on the patent technology information using an image generation model;

[0042] A video generation module, the module is used to receive the information for generating text keywords and a request for generating text information, audio information and video information, extract text keywords from the information for generating text keywords using a natural language processing model, generate text information, generate audio information based on the text information using a text-to-speech technology, and combine the audio information, text information and picture information using a video editing software to synthesize video information containing the text information and the audio information;

[0043] An interaction and feedback module is used to respond and interact with the front-end interface unit and feed back the generated picture information, text information, audio information and video information to the front-end interface unit.

[0044] Preferably, the database unit includes a user data storage module and a video data storage module; wherein:

[0045] The user data storage module is used to store the patent technology information and information used to generate copywriting keywords;

[0046] The video data storage module is used to store the picture information, text information, audio information and video information.

[0047] According to a third aspect of the present invention, there is provided a computer conversion device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, can be used to execute any of the methods described above in the present invention, or to run any of the systems described above in the present invention.

[0048] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can be used to execute any of the methods described above in the present invention, or to run any of the systems described above in the present invention.

[0049] Due to the adoption of the above technical solution, the present invention has at least one of the following beneficial effects compared with the prior art:

[0050] The method and system for converting patent text into video based on AI technology provided by the present invention can automatically generate video content from obscure patent technical documents, greatly shortening the video production cycle and improving production efficiency. At the same time, it is beneficial for users to intuitively understand the technical content, thereby facilitating the transformation of scientific and technological achievements and better promoting and popularizing patent technology.

[0051] The method and system for converting patent text into video based on AI technology provided by the present invention ensure the high correlation between the generated image and the text, as well as the accurate coherence and knowledge expertise between the technical content in the video, by accurately parsing the keywords in the patent technology, thereby improving the quality of the generated video and reducing the cost investment in the manual video production process.

[0052] The method and system for converting patent text into video based on AI technology provided by the present invention automatically integrates text, pictures and dubbing into high-quality video content through the coordination of multiple media elements, providing users with a richer and more intuitive visual experience; in addition, for the visually impaired, the generated video content can also help them better understand and learn related technologies.

[0053] The method and system for converting patent text into video based on AI technology provided by the present invention can quickly generate promotional videos that are consistent with patent technology according to user needs, thereby achieving efficient technology dissemination; at the same time, it can also adjust the video generation process in real time according to user feedback to ensure that the generated video meets user expectations and can be easily adapted to a variety of media platforms, which enables the promotion of patent technology to cover a wider audience and improve the popularity and influence of the technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0055] Figure 1 The present invention is a flowchart of a method for converting patent text into video based on AI technology in a preferred embodiment of the present invention.

[0056] Figure 2 This is a schematic diagram of the composition architecture and working content of a system for converting patent text into video based on AI technology in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following is a detailed description of the embodiments of the present invention: This embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation method and a specific operation process are given. It should be pointed out that for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

[0058] As an intuitive and vivid information carrier, video helps people to intuitively understand the technical content, thereby facilitating the transformation of scientific and technological achievements. However, due to the complexity of the technical information contained in patent technology and the high degree of professional knowledge, so far, no technology has been found to convert scientific and technological achievements and / or patent documents into videos for presentation through the method of Wensheng video.

[0059] In response to the above-mentioned technical problems and technical bottlenecks in the prior art, an embodiment of the present invention provides a system for converting patent texts into videos based on AI technology. The system automatically generates video content with clear logic, fluent language, and meeting user expectations by extracting relevant patent information (such as patent name, composition, efficacy, etc.) from the patent technology text content and / or pictures provided by the user.

[0060] Specifically, Figure 1 As shown, the method for converting patent text into video based on AI technology provided in this embodiment may include the following operations:

[0061] S1, obtain patent technology information and information used to generate copywriting keywords;

[0062] S2, generating picture information according to the patent technology information, generating text information and audio information according to the information for generating text keywords, and synthesizing the picture information, text information and audio information to obtain video information;

[0063] S3, adjusting the patent technology information and / or the information used to generate the copy keywords according to the generated image information, copy information, audio information and video information, and repeating the previous step until the video conversion of the patent technology is completed.

[0064] In some preferred implementations, in the above S1, the patent technology information may include: patent technology text data, patent technology image data, etc.

[0065] In some preferred implementations, in the above S1, the information used to generate copy keywords may include: keyword information set by the user, patent technology information, etc.

[0066] In order to obtain the picture elements in the video information, in some preferred implementations, in the above S2, generating the picture information according to the patent technology information may further include the following operations:

[0067] S21, preprocessing the patent technical information to obtain key graphic information and key technical text information including semantic structure;

[0068] S22, constructing a visual element library using the key graphic information;

[0069] S23, providing a patent technology information keyword library and a natural language processing model, and using the patent technology information keyword library to train the natural language processing model to obtain a keyword extraction model;

[0070] S24, taking the key technical text information as input of the keyword extraction model, and using the understanding of semantics by the keyword extraction model to obtain technical keywords of the key technical text information;

[0071] S25, selecting key graphic information matching the technical keyword from the visual element library according to the result of semantic understanding, to form a sample to be generated;

[0072] S26, providing an image generation model, using the sample to be generated as input of the image generation model, and obtaining image information that is highly relevant to the patent technology information.

[0073] In order to improve the image quality, the obtained image information may be further optimized, which may include any one or more of the following operations:

[0074] Remove unnecessary parts of the image;

[0075] Adjust the color distribution, color saturation, hue and contrast in the picture to make the core content of the picture more prominent;

[0076] Sharpen the edge details of the image;

[0077] Create a high-frequency layer containing only the edges of the image and overlay it on the original image to enhance the sharpness of the image;

[0078] Remove noise from the image and improve the clarity of the image;

[0079] Add watermark to your images to protect their copyright.

[0080] In S26, the image generation model may be a GAN, a VAE, a diffusion model, or the like.

[0081] In order to obtain accurate and coherent technical solutions in the patent technical information, in some preferred implementations, the preprocessing of the patent technical information in S21 may further include the following operations:

[0082] S211, obtaining patent technical information in text format;

[0083] S212, extracting key graphic information from the patent technical information in text format for building a visual element library;

[0084] S213, preprocessing the patent technical information in text format, which may include word segmentation, part-of-speech tagging, syntactic analysis and other processing on the patent technical information in text format, to obtain key technical text information and semantic structure of the patent technical information in text format.

[0085] In order to obtain the text elements in the video information, in some preferred implementations, in the above S2, the step of generating the text information according to the information used to generate text keywords may further include the following operations:

[0086] M21, providing a patent technology information keyword library and a natural language processing model, and using the patent technology information keyword library to train the natural language processing model to obtain a keyword extraction model;

[0087] M22, using the keyword extraction model to understand the semantics of the information used to generate copy keywords, identifying copy keywords in the information used to generate copy keywords, and then generating copy information.

[0088] In order to obtain accurate copywriting keywords, in S2, expert systems can be used in combination with prior knowledge with semantic structure to process and analyze a large number of patent documents, obtain technical keywords and build a patent technical information keyword library for training natural language processing models. At the same time, the following technical means can also be used to expand and update the patent technical information keyword library:

[0089] Obtain patent documents to construct a file set, and perform pre-processing on the documents such as part-of-speech tagging, syntactic analysis, and semantic role tagging;

[0090] A keyword extraction algorithm is used to extract technical keywords from patent documents; wherein the keyword extraction algorithm may include:

[0091] The TF-IDF algorithm calculates the importance of a keyword based on its frequency of appearance in a document and in a document set, thereby obtaining the required technical keywords.

[0092] TextRank algorithm: Based on the PageRank algorithm in graph theory, it uses the co-occurrence relationship between keywords to build a graph, and calculates the importance of keywords based on the weights between keywords to obtain the required technical keywords;

[0093] LDA algorithm: discover topics based on the vocabulary distribution in the file; obtain the topic distribution in each file through the LDA algorithm, and then calculate the importance of each keyword based on the topic distribution, and finally obtain the required technical keywords;

[0094] RAKE algorithm: focuses on keyword co-occurrence and position, calculates the score of each candidate keyword based on the co-occurrence relationship between keywords and the position of keywords in the text, and finally obtains the required technical keywords.

[0095] In S2, the natural language processing model can be GPT-4.

[0096] In some preferred implementations, in the above M22, the copy keywords may include: elements related to the theme, concept and / or relationship.

[0097] Furthermore, with respect to patent technical information, the subject corresponds to the patent name and technical effect in the patent technical information, the concept corresponds to the technical indicators in the patent technical information, and the relationship corresponds to the technical field and / or application scenario in the patent technical information, so that the copywriting keywords are highly relevant to the patent technical information.

[0098] In order to obtain audio elements in the video information, in some preferred implementations, in the above S2, audio information is generated according to the information used to generate text keywords, and text-to-speech (TTS) technology can be further used to generate audio information (dubbing) based on the text information.

[0099] In order to obtain the final video, in some preferred implementations, in the above S3, the step of synthesizing the picture information, text information and audio information to obtain the video information may further include the following operations:

[0100] Computer vision technology is used to arrange the picture information, text information and audio information, add animation effects, and match sound effects and background music, ultimately generating patented technology video information with technical points and visual effects.

[0101] In order to improve the video quality, the above S3 may further include the following operations:

[0102] Video editing tools are used to edit and process special effects on the obtained video information.

[0103] In S3, computer vision techniques can be implemented using video editing tools and libraries, such as FFmpeg, OpenCV, and FreeImage.

[0104] In S2, the process of generating image information first requires obtaining patent technical information in text format. Patent technical information in the form of patent technical text data can be directly used as patent technical information in text format; patent technical information in the form of patent technical image data needs to be converted from image to text technology (such as OCR text recognition technology) to obtain patent technical information in text format.

[0105] In S2, the keyword information set by the user is used as the information for generating the copy keywords, which may be ambiguous or ambiguous. The embodiment of the present invention can achieve semantic understanding of the keyword information set by the user through a natural language processing model pre-trained with a patent technology information keyword library, and then identify accurate and coherent copy keywords.

[0106] In order to better promote and popularize patented technologies, the above-mentioned embodiments of the present invention automatically generate advertising videos based on the provided patented technology text content and / or image content, as well as related copy keywords (for example: themes, indicators, application scenarios, effects, etc.).

[0107] The above-mentioned embodiment of the present invention adopts multimodal fusion technology to fuse multiple algorithm models to generate various elements, combines the generated picture elements, text elements and audio elements, automatically generates an advertising video, and uses video editing tools to edit and process the video with special effects to obtain a high-quality patent technology video that is highly consistent with the patent technology information.

[0108] in:

[0109] Multimodal fusion technology is used to integrate multiple algorithm models, specifically:

[0110] A collection of various deep learning models, including image generation models and natural language recognition models, respectively obtain image information and text information to improve the realism, detail richness, accuracy and generation efficiency of video generation.

[0111] Semantic understanding and content recognition automatically generate video content that meets the requirements based on input text or text description.

[0112] Visual fusion technology enables seamless integration of multi-source information such as audio, text, and images, enhancing video expressiveness.

[0113] Automated editing and post-processing, through intelligent editing technology, improve the automation level and final quality of video production.

[0114] The above embodiments of the present invention solve the following technical problems and achieve the following technical effects:

[0115] 1. About keyword analysis and copywriting generation

[0116] The keywords input by the user and the keywords in the obtained patent technical information may be ambiguous or polysemic, and it is difficult to parse these keywords to generate accurate and coherent copy. The above embodiment of the present invention uses a natural language processing model pre-trained with a patent technical information keyword library to achieve semantic understanding of the keyword information set by the user, and then identify accurate and coherent copy keywords, thereby achieving precise keyword parsing, so that the obtained copy information is highly relevant to the technical solutions in the patent technical information and conforms to professional knowledge, thereby ensuring the quality of the copy.

[0117] 2. About image generation and processing

[0118] In the process of generating image information, we first obtain the patent technical information in text format, and then obtain the key graphic information and key technical text information containing semantic structure. Then, we use the natural language processing model to understand the semantics, extract technical keywords, and match the graphic information. Finally, we use the image generation model to obtain image information that is highly relevant to the patent technical information, ensuring the consistency between the image content and the high correlation with the technical solution.

[0119] In addition, the obtained image information can be further optimized to improve the visual effect.

[0120] 3. About video generation and editing

[0121] By using computer vision technology, the generated image information, text information and audio information are arranged, animation effects are added, and sound effects and background music are matched. The final video information not only has technical points and visual effects, but also meets customer requirements.

[0122] IV. User Interaction and Feedback

[0123] Users can also adjust patent technology information and / or information used to generate copy keywords based on the generated video information to ensure that the input keywords and patent technology information meet user requirements, and then adjust the video generation process in real time to ensure that the generated video meets user expectations.

[0124] Based on the same inventive concept, an embodiment of the present invention also provides a system for converting patent text into video based on AI technology.

[0125] Specifically, Figure 2 As shown, the system for converting patent text into video based on AI technology provided in this embodiment may include the following functional units:

[0126] A front-end interface unit, which is used to input patent technology information and information for generating copy keywords, and receive and generate image information, copy information, audio information and video information, and / or display video information;

[0127] A back-end server unit, which is used to generate image information according to the patent technology information, generate text information and audio information according to the information used to generate text keywords, synthesize the image information, text information and audio information to obtain video information, and feed it back to the front-end interface unit;

[0128] A database unit is used to store the patent technology information and information used to generate copy keywords as well as the picture information, copy information, audio information and video information.

[0129] In some preferred implementations, the front-end interface unit may further include the following modules:

[0130] A user upload module, which is used to input patent technology information and upload it to the back-end server unit, and at the same time output a request to generate image information of the patent technology information to the back-end server unit;

[0131] A copywriting generation module, the module is used to input information for generating copywriting keywords and upload it to the back-end server unit, and at the same time output a request to generate copywriting information, audio information and video information for the information for generating copywriting keywords to the back-end server unit;

[0132] A feedback and interaction module, which is used to provide an interactive interface and use the received image information, text information, audio information and video information to adjust the input patent technology information and / or information used to generate text keywords;

[0133] The display module is used to display the received video information.

[0134] In some preferred implementations, the backend server unit may further include the following operations:

[0135] An image generation module, the module is used to receive the patent technology information and a request for generating image information, and generate image information based on the patent technology information using an image generation model;

[0136] A video generation module, the module is used to receive the information for generating text keywords and a request for generating text information, audio information and video information, extract text keywords from the information for generating text keywords using a natural language processing model, generate text information, generate audio information based on the text information using a text-to-speech (TTS) technology, and combine the audio information, text information and picture information using a video editing software to synthesize video information containing the text information and the audio information;

[0137] An interaction and feedback module is used to respond and interact with the front-end interface unit and feed back the generated picture information, text information, audio information and video information to the front-end interface unit.

[0138] In some preferred implementations, the above database unit may further include: a user data storage module and a video data storage module; wherein:

[0139] The user data storage module is used to store the patent technology information and information used to generate copywriting keywords;

[0140] The video data storage module is used to store the picture information, text information, audio information and video information.

[0141] The system for converting patent text into video based on AI technology provided by the above-mentioned embodiment of the present invention realizes intelligent interaction and real-time feedback through the front-end interface unit and the back-end server unit, ensuring that users can easily enter keywords and provide patent technical information files, and obtain timely feedback, and adjust the generated video information in real time according to the feedback to ensure the accuracy and professionalism of the video.

[0142] It should be noted that the steps in the method provided by the present invention can be implemented by using corresponding modules, devices, units, etc. in the system. Those skilled in the art can refer to the technical solution of the system to implement the step flow of the method, that is, the embodiments in the system can be understood as preferred examples of implementing the method, which will not be elaborated here.

[0143] An embodiment of the present invention also provides a computer conversion device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it can be used to execute any method of the above-mentioned embodiments of the present invention, or to run any system of the above-mentioned embodiments of the present invention.

[0144] Optionally, the memory is used to store programs; the memory may include volatile memory (English: volatile memory), such as random-access memory (English: random-access memory, abbreviated: RAM), such as static random-access memory (English: static random-access memory, abbreviated: SRAM), double data rate synchronous dynamic random access memory (English: Double Data Rate Synchronous Dynamic Random Access Memory, abbreviated: DDR SDRAM), etc.; the memory may also include non-volatile memory (English: non-volatile memory), such as flash memory (English: flash memory). The memory is used to store computer programs (such as applications, functional modules, etc. that implement the above method), computer instructions, etc., and the above computer programs, computer instructions, etc. can be partitioned and stored in one or more memories. And the above computer programs, computer instructions, data, etc. can be called by the processor.

[0145] The above-mentioned computer programs, computer instructions, etc. may be stored in one or more memories in partitions, and the above-mentioned computer programs, computer instructions, data, etc. may be called by a processor.

[0146] The processor is used to execute the computer program stored in the memory to implement the various steps of the method or various modules of the system involved in the above embodiments. For details, please refer to the relevant descriptions in the above method and system embodiments.

[0147] The processor and the memory may be independent structures or integrated structures. When the processor and the memory are independent structures, the memory and the processor may be coupled and connected via a bus.

[0148] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it can be used to run the system of any one of the above embodiments of the present invention, or to execute the method of any one of the above embodiments of the present invention.

[0149] Among them, computer-readable media include computer storage media and communication media, wherein the communication media include any media that facilitates the transmission of computer programs from one place to another. The storage medium can be any available medium that can be accessed by a general or special-purpose computer. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be an integral part of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in a user device. Of course, the processor and the storage medium can also be present in a communication device as discrete components.

[0150] The system and method for converting patent text into video based on AI technology provided by the above-mentioned embodiment of the present invention can automatically generate video content from obscure patent technical documents, greatly shorten the video production cycle, improve production efficiency, and at the same time help users intuitively understand the technical content, thereby facilitating the transformation of scientific and technological achievements, and better promoting and popularizing patent technology; by accurately parsing the keywords in the patent technology, the high correlation between the generated image and the text and the accurate coherence and knowledge professionalism between the technical content in the video are guaranteed, the quality of the generated video is improved, and the cost investment in the manual video production link is reduced; by coordinating multiple media elements, the text, pictures and dubbing are automatically integrated into high-quality video content, providing users with a richer and more intuitive visual body; in addition, for the visually impaired, the generated video content can also help them better understand and learn related technologies; according to user needs, promotional videos that match the patent technology can be quickly generated to achieve efficient technology dissemination; at the same time, the video generation process can be adjusted in real time according to user feedback to ensure that the generated video meets user expectations and can easily adapt to multiple media platforms, such as social media, websites, blogs, etc., which enables the promotion of patent technology to cover a wider audience and improve the popularity and influence of the technology.

[0151] All matters not covered in the above embodiments of the present invention are well known in the art.

[0152] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art may make various modifications or variations within the scope of the claims, which do not affect the essence of the present invention.

Claims

1. A method for converting patent text into video based on AI technology, characterized in that: include: Obtain patent technology information and information used to generate copywriting keywords; Generate image information according to the patent technology information, generate text information and audio information according to the information for generating text keywords, and synthesize the image information, text information and audio information to obtain video information; According to the generated image information, text information, audio information and video information, adjust the patent technology information and / or the information used to generate text keywords, and repeat the previous step until the video conversion of the patent technology is completed.

2. The method for converting patent text into video based on AI technology according to claim 1, characterized in that: The patent technology information includes: patent technology text data and / or patent technology image data; and / or The information used to generate copy keywords includes: keyword information and / or patent technology information set by the user.

3. The method for converting patent text into video based on AI technology according to claim 1, characterized in that: The generating of image information according to the patent technology information includes: Preprocess the patented technical information to obtain key graphic information and key technical text information containing semantic structure; Using the key graphic information to construct a visual element library; Providing a patent technology information keyword library and a natural language processing model, using the patent technology information keyword library to train the natural language processing model to obtain a keyword extraction model; Using the key technical text information as input to the keyword extraction model, and using the keyword extraction model's understanding of semantics to obtain technical keywords of the key technical text information; According to the result of semantic understanding, key graphic and text information matching the technical keywords is selected from the visual element library to form a sample to be generated; An image generation model is provided, and the sample to be generated is used as the input of the image generation model to obtain image information that is highly relevant to the patent technology information.

4. The method for converting patent text into video based on AI technology according to claim 3, characterized in that: The preprocessing of the patent technology information includes: Obtain patent technology information in text format; Extracting key graphic information from the patent technical information in text format for building a visual element library; The patent technical information in the text format is preprocessed to obtain key technical text information and semantic structure of the patent technical information in the text format.

5. The method for converting patent text into video based on AI technology according to claim 1, characterized in that: Generating copywriting information according to the information for generating copywriting keywords includes: Providing a patent technology information keyword library and a natural language processing model, using the patent technology information keyword library to train the natural language processing model to obtain a keyword extraction model; The keyword extraction model is used to understand the semantics of the information used to generate the copywriting keywords, identify the copywriting keywords in the information used to generate the copywriting keywords, and then generate the copywriting information.

6. The method for converting patent text into video based on AI technology according to claim 5, characterized in that: The copywriting keywords include elements related to the theme, concept and / or relationship; among which: The subject corresponds to the patent name and technical effect in the patent technical information, the concept corresponds to the technical indicators in the patent technical information, and the relationship corresponds to the technical field and / or application scenario in the patent technical information.

7. The method for converting patent text into video based on AI technology according to claim 1, characterized in that: The audio information is generated according to the information used to generate the copy keywords, and the audio information is generated based on the copy information using text-to-speech technology.

8. The method for converting patent text into video based on AI technology according to claim 1, characterized in that: The synthesizing the picture information, text information and audio information to obtain video information includes: Computer vision technology is used to arrange the picture information, text information and audio information, add animation effects, and match sound effects and background music, ultimately generating patented technology video information with technical points and visual effects.

9. A system for converting patent text into video based on AI technology, characterized in that: include: A front-end interface unit, which is used to input patent technology information and information for generating copy keywords, and receive and generate image information, copy information, audio information and video information, and / or display video information; A back-end server unit, which is used to generate image information according to the patent technology information, generate text information and audio information according to the information used to generate text keywords, synthesize the image information, text information and audio information to obtain video information, and feed it back to the front-end interface unit; A database unit is used to store the patent technology information and information for generating copy keywords as well as the picture information, copy information, audio information and video information.

10. The system for converting patent text into video based on AI technology according to claim 9, characterized in that: The front-end interface unit comprises: A user upload module, which is used to input patent technology information and upload it to the back-end server unit, and at the same time output a request to generate image information of the patent technology information to the back-end server unit; A copywriting generation module, the module is used to input information for generating copywriting keywords and upload it to the back-end server unit, and at the same time output a request to generate copywriting information, audio information and video information for the information for generating copywriting keywords to the back-end server unit; A feedback and interaction module, which is used to provide an interactive interface and use the received image information, text information, audio information and video information to adjust the input patent technology information and / or information used to generate text keywords; The display module is used to display the received video information.

11. The system for converting patent text into video based on AI technology according to claim 9, characterized in that: The back-end server unit comprises: An image generation module, the module is used to receive the patent technology information and a request for generating image information, and generate image information based on the patent technology information using an image generation model; A video generation module, the module is used to receive the information for generating text keywords and a request for generating text information, audio information and video information, extract text keywords from the information for generating text keywords using a natural language processing model, generate text information, generate audio information based on the text information using a text-to-speech technology, and combine the audio information, text information and picture information using a video editing software to synthesize video information containing the text information and the audio information; An interaction and feedback module is used to respond and interact with the front-end interface unit and feed back the generated picture information, text information, audio information and video information to the front-end interface unit.

12. The system for converting patent text into video based on AI technology according to claim 9, characterized in that: The database unit includes a user data storage module and a video data storage module; wherein: The user data storage module is used to store the patent technology information and information used to generate copywriting keywords; The video data storage module is used to store the picture information, text information, audio information and video information.

13. A computer conversion device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it can be used to perform the method described in any one of claims 1 to 8, or to run the system described in any one of claims 9 to 12.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it can be used to run the method described in any one of claims 1 to 8, or to run the system described in any one of claims 9 to 12.

Images