New media content creation and distribution platform

Abstract

The invention discloses a new media content creation and distribution platform, and the platform comprises a user portrait construction module which is used for collecting user behavior data and generating a multi-dimensional interest label; the multi-modal content generation module supports automatic generation and editing of texts, images, audios and videos; an advanced algorithm is integrated through a multi-modal content generation module, intelligent collaborative production of texts, images, audios and videos is achieved, creation efficiency is improved, title variant generation and style imitation are supported, and the manual creative bottleneck is broken through; the intelligent distribution engine adopts reinforcement learning to dynamically optimize weight distribution and combines a platform feature analyzer and an A / B test unit to improve cross-platform distribution accuracy and traffic conversion rate; the block chain evidence storage and cross-platform crawler system constructs a full-link copyright protection network, so that content fingerprints can be quickly chained, infringement contents can be automatically released, and the right protection efficiency is improved; and the virtual digital human driving module and the multi-mode conversion unit extend the content adaptability, so that the immersion of the user is enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of new media content creation and distribution, and particularly relates to a new media content creation and distribution platform. Background Art

[0002] The new media industry currently faces multiple technical challenges: Traditional content production relies on manual labor, with insufficient tool intelligence, resulting in low creation efficiency and difficulty in scaling up; Cross-platform distribution requires manual adaptation of different formats, which is time-consuming and error-prone; Content homogenization is serious, lacking the ability of personalized generation; The copyright protection mechanism is weak, and it is difficult to track infringement.

[0003] In the prior art, content generation tools are mostly limited to single modalities, lacking the ability of multi-modal collaborative generation and unable to meet the needs of composite forms. In the distribution link, most platforms adopt static recommendation strategies, failing to dynamically combine user behavior with platform characteristics, resulting in low content reach rate. In addition, existing copyright management systems mostly rely on centralized deposit of evidence, with the risk of tampering, and the coverage of cross-platform infringement monitoring is insufficient. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the existing defects and provide a new media content creation and distribution platform to solve the problems raised in the above background art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A new media content creation and distribution platform, including:

[0006] A user portrait construction module, used to collect user behavior data and generate multi-dimensional interest tags;

[0007] A multi-modal content generation module, supporting the automatic generation and editing of text, images, audio, and video;

[0008] An intelligent distribution engine, dynamically adjusting the distribution strategy according to platform characteristics;

[0009] A copyright management module, realizing content traceability and infringement monitoring;

[0010] A data analysis center, providing real-time tracking of dissemination effects and suggestions for strategy optimization.

[0011] Preferably, the multi-modal content generation module includes:

[0012] A natural language processing unit, realizing the generation of hot texts based on the Transformer architecture;

[0013] An image synthesis unit, supporting the generation of high-resolution matching pictures according to text descriptions;

[0014] A video editing unit, automatically completing shot switching, transition effects, and background music matching;

[0015] A cross-modal conversion unit that converts text content into the form of comics, podcasts, or short videos.

[0016] Preferably, the natural language processing unit includes:

[0017] A creative title generator that uses a Bi-LSTM model to generate more than 20 variant titles;

[0018] A sensitive word filtering component that implements real-time content compliance review based on a dynamic word library;

[0019] A style transferrer that can create content by imitating the copywriting style of a specified KOL.

[0020] Preferably, the intelligent distribution engine includes:

[0021] A platform feature analyzer that extracts the format specifications, user active periods, and content preferences of each social media platform;

[0022] A dynamic weight allocator that optimizes the distribution priority using a reinforcement learning algorithm;

[0023] A real-time bidding unit that connects to an advertising trading platform for precise traffic procurement;

[0024] An A / B testing unit that concurrently tests multiple distribution schemes and automatically selects the optimal solution.

[0025] Preferably, the copyright management module implements:

[0026] Content fingerprinting and archiving based on blockchain, generating a SHA-256 hash value and uploading it to the blockchain;

[0027] A cross-platform content crawler system that regularly scans for infringing content;

[0028] An automatic rights protection subsystem that generates an infringement report and initiates a request for removal.

[0029] Preferably, the user profile construction module includes:

[0030] A social relationship graph generator that analyzes the user's follow, like, and share relationship chains;

[0031] An interest prediction model that predicts content preferences based on the user's viewing duration and completion rate;

[0032] A device feature analyzer that adapts the content format according to the terminal type and network environment.

[0033] Preferably, it further includes:

[0034] A virtual digital human driving module that generates a virtual anchor broadcast video based on the user's input text;

[0035] Intelligent voice dubbing unit, providing more than 50 voice styles and multi-language conversion;

[0036] Expression and action library, containing more than 2000 preset virtual human action templates.

[0037] Preferably, the data analysis center provides:

[0038] Propagation path visualization tool, drawing the diffusion map of content in the social network;

[0039] ROI calculator, counting the input-output ratio of a single piece of content;

[0040] Intelligent review report generator, automatically outputting a list of content optimization suggestions.

[0041] A distribution method for a new media content creation and distribution platform, including the following steps:

[0042] S1: Extract content keywords and sentiment tendencies through semantic analysis;

[0043] S2: Match the historical popular content feature library to find similar patterns;

[0044] S3: Use knowledge graph technology to construct a triangular relationship model of platform - user - content;

[0045] S4: Generate a distribution plan including release time, platform combination, and promotion budget;

[0046] S5: Dynamically adjust the distribution strategy according to real-time feedback data.

[0047] Compared with the prior art, the present invention provides a new media content creation and distribution platform, having the following beneficial effects:

[0048] The present invention integrates advanced algorithms through a multi-modal content generation module to realize the intelligent collaborative production of text, images, audio and video, improve the creation efficiency, support the generation of title variants and style imitation, and break through the artificial creativity bottleneck; the intelligent distribution engine uses reinforcement learning to dynamically optimize weight allocation, combines the platform feature analyzer and the A / B test unit to improve the cross-platform distribution accuracy and traffic conversion rate; the blockchain evidence storage and cross-platform crawler system constructs a full-link copyright protection network, realizes the rapid uploading of content fingerprints and the automatic removal of infringing content, and improves the rights protection efficiency; the virtual digital human driving module and the multi-modal conversion unit expand the content adaptability and enhance the user immersion. Detailed implementation mode

[0049] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Apparently, the described embodiments are only a part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0050] The present invention provides a technical solution: a new media content creation and distribution platform, including:

[0051] A user portrait construction module 110, configured to collect user behavior data and generate multi-dimensional interest tags;

[0052] A multi-modal content generation module 120, which supports the automatic generation and editing of text, images, audio, and videos;

[0053] An intelligent distribution engine 130, which dynamically adjusts the distribution strategy according to the platform characteristics;

[0054] A copyright management module 140, which realizes content traceability and infringement monitoring;

[0055] A data analysis center 150, which provides real-time tracking of the dissemination effect and suggestions for strategy optimization.

[0056] In the present invention, preferably, the multi-modal content generation module 120 includes:

[0057] A natural language processing unit 121, which realizes the generation of hot texts based on the Transformer architecture;

[0058] An image synthesis unit 122, which supports the generation of high-resolution matching pictures according to text descriptions;

[0059] A video editing unit 123, which automatically completes shot switching, transition effects, and background music matching;

[0060] A cross-modal conversion unit 124, which converts text content into forms such as comics, podcasts, or short videos.

[0061] In the present invention, preferably, the natural language processing unit 121 includes:

[0062] A creative title generator, which uses a Bi-LSTM model to generate more than 20 variant titles;

[0063] A sensitive word filtering component, which realizes real-time content compliance review based on a dynamic word library;

[0064] A style transferrer, which can imitate the copywriting style of a specified KOL for creation.

[0065] In the present invention, preferably, the intelligent distribution engine 130 includes:

[0066] The platform feature analyzer 131 extracts the format specifications, user active periods, and content preferences of each social media platform;

[0067] The dynamic weight allocator 132 optimizes the distribution priority using a reinforcement learning algorithm;

[0068] The real-time bidding unit 133 connects to the advertising trading platform for precise traffic procurement;

[0069] The A / B test unit 134 concurrently tests multiple distribution schemes and automatically selects the optimal solution.

[0070] In the present invention, preferably, the copyright management module 140 realizes:

[0071] Content fingerprint storage based on blockchain, generating a SHA-256 hash value and uploading it to the blockchain;

[0072] A cross-platform content crawler system that regularly scans for infringing content;

[0073] An automatic rights protection subsystem that generates an infringement report and initiates a take-down request.

[0074] In the present invention, preferably, the user portrait construction module 110 includes:

[0075] The social relationship graph generator 111 analyzes the user's follow, like, and share relationship chains;

[0076] The interest prediction model 112 predicts content preferences based on the user's viewing duration and completion rate;

[0077] The device feature analyzer 113 adapts the content format according to the terminal type and network environment.

[0078] In the present invention, preferably, it further includes:

[0079] The virtual digital human driving module 160 generates a virtual anchor broadcast video based on the user's input text;

[0080] The intelligent voice dubbing unit 161 provides more than 50 voice styles and multi-language conversion;

[0081] The expression and action library 162 contains more than 2000 pre-set virtual human action templates.

[0082] In the present invention, preferably, the data analysis center 150 provides:

[0083] The dissemination path visualization tool 151 draws the diffusion map of the content in the social network;

[0084] The ROI calculator 152 calculates the input-output ratio of a single piece of content;

[0085] The intelligent review report generator 153 automatically outputs a list of content optimization suggestions.

[0086] A distribution method for a new media content creation and distribution platform, comprising the following steps:

[0087] S1: Extract content keywords and sentiment tendencies through semantic analysis;

[0088] S2: Match the historical popular content feature library to find similar patterns;

[0089] S3: Use knowledge graph technology to construct a triangular relationship model of platform - user - content;

[0090] S4: Generate a distribution plan including release time, platform combination, and promotion budget;

[0091] S5: Dynamically adjust the distribution strategy according to real - time feedback data.

[0092] Example 1:

[0093] A new media content creation and distribution platform, comprising the following steps:

[0094] Step 1, content generation; Input the keyword "beauty product review", the natural language processing unit 121 automatically generates 20 groups of creative titles (such as "The top - selling products on the whole network! These 5 makeup products are a must - buy without any pitfalls"); The image synthesis unit 122 generates 3D product display pictures according to the product description, and the video editing unit 123 calls the preset transition template to automatically splice the product trial materials; The cross - modal conversion unit 124 converts the review copy into a 15 - second vertical short video + graphic note dual form;

[0095] Step 2, intelligent distribution; The platform feature analyzer 131 identifies that the active period of Douyin users is 18:00 - 22:00, and Xiaohongshu users prefer step - by - step content; The dynamic weight allocator 132 calculates that the weight of Douyin is 0.6 (high traffic), the weight of Bilibili is 0.3 (long - tail effect), and the weight of Kuaishou is 0.1; The A / B test unit 134 simultaneously launches 3 cover picture schemes, and selects the "lipstick swatch comparison picture" with the highest click - through rate after 2 hours;

[0096] Step 3, copyright protection; The blockchain evidence storage system generates a video fingerprint (SHA - 256), and after detecting unauthorized reproduction on a certain platform, the automatic rights protection subsystem completes the removal within 24 hours;

[0097] Technical effects: The creation time is shortened from the traditional 8 hours to 2.5 hours; The Douyin play volume exceeds 5 million times, and the conversion rate of Xiaohongshu is increased to 12%; The recognition accuracy of infringing content reaches 98%, and the rights protection response time is shortened by 80%.

[0098] Example 2:

[0099] A new media content creation and distribution platform includes the following steps:

[0100] Step 1: Multimodal conversion: The original article is adjusted through the style transfer tool into: WeChat long picture and text (in-depth interpretation), Weibo topic (new product spoilers + nine-square grid image), and headline information flow (data visualization chart); the image synthesis unit 122 automatically extracts key data to generate information graphics, and the sensitive word filtering component blocks advertising banned words such as "exclusive" and "preferred" in real time;

[0101] Step 2: Dynamic Distribution: The user profile building module 110 analyzes Weibo users' preferences for entertainment content and converts the core selling points of the article into celebrity recommendations. The real-time bidding unit 133 targets female users aged 25-35 in the WeChat Moments ad bidding, reducing the CPM cost to 70% of the industry average.

[0102] Step 3: Tracking the impact: The communication path visualization tool 151 shows that the article has been recirculated in the maternity and baby community through KOL forwarding; the ROI calculator 152 shows that the cost of acquiring a single piece of content has decreased by 45% compared to the previous month;

[0103] Technical effect: The time required for cross-platform adaptation was reduced from 4 hours of manual operation to automatic completion; the number of readings of Weibo topics reached 230 million times, and the WeChat open rate increased to 28%; the intelligent review report generated optimization suggestions: increase the weight of short video unboxing content.

[0104] Example 3:

[0105] A new media content creation and distribution platform includes the following steps:

[0106] Step 1: Content Production: Input the product parameter table, and the virtual digital human driver module 160 automatically generates a bilingual live broadcast script in English and Thai. The expression and action library 162 matches demonstration actions based on product characteristics (such as calling the "makeup trial close-up shot" template for beauty products). The intelligent dubbing unit 161 synthesizes American English and Thai voices, and the speaking speed automatically adapts to the recommended values of each platform (YouTube 150 words / minute, TikTok 180 words / minute).

[0107] Step 2: Intelligent distribution: The device feature analyzer 113 automatically downgrades the video resolution to 720P for Southeast Asian users to adapt to mobile networks; the dynamic weight allocator 132 concentrates 80% of the budget on the TikTok live broadcast room during the Prime Time period (20:00-22:00);

[0108] Step 3: Copyright management: Live broadcasts and recordings generate content fingerprints in real time. If a platform detects illegal broadcasts, the system automatically triggers a DMCA complaint and replaces the content with a pre-set copyright protection watermark.

[0109] Technical effects: The preparation time for a single live broadcast is compressed from 3 days to 6 hours; the viewing duration in the Southeast Asian market is increased to an average of 18 minutes per person; the discovery time limit for piracy is shortened from 72 hours to 3 hours.

[0110] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A new media content creation and distribution platform, characterized in that, Including: A user profile construction module (110) for collecting user behavior data and generating multi-dimensional interest tags; A multi-modal content generation module (120) that supports the automated generation and editing of text, images, audio, and video; An intelligent distribution engine (130) that dynamically adjusts the distribution strategy according to platform characteristics; A copyright management module (140) that realizes content traceability and infringement monitoring; A data analysis center (150) that provides real-time dissemination effect tracking and strategic optimization suggestions.

2. The new media content creation and distribution platform according to claim 1, characterized in that The multi-modal content generation module (120) includes: A natural language processing unit (121) that realizes the generation of hot text based on the Transformer architecture; An image synthesis unit (122) that supports the generation of high-resolution illustrations according to text descriptions; A video editing unit (123) that automatically completes shot transitions, transition effects, and background music matching; A cross-modal conversion unit (124) that converts text content into comic, podcast, or short video forms.

3. A new media content creation and distribution platform according to claim 2, characterized in that, The natural language processing unit (121) includes: A creative title generator that uses a Bi-LSTM model to generate more than 20 variant titles; A sensitive word filtering component that realizes real-time content compliance review based on a dynamic word library; A style transferrer that can imitate the copywriting style of a specified KOL for creation.

4. A new media content creation and distribution platform according to claim 1, characterized in that The intelligent distribution engine (130) includes: A platform feature analyzer (131) that extracts the format specifications, user active periods, and content preferences of each social media platform; A dynamic weight allocator (132) that uses a reinforcement learning algorithm to optimize the distribution priority; A real-time bidding unit (133) that docks with an advertising trading platform for precise traffic procurement; An A / B test unit (134) that parallelly tests multiple distribution schemes and automatically selects the optimal solution.

5. A new media content creation and distribution platform according to claim 1, characterized in that, The copyright management module (140) realizes: Content fingerprinting and archiving based on blockchain, generating a SHA-256 hash value and uploading it to the blockchain; A cross-platform content crawler system that regularly scans for infringing content; An automatic rights protection subsystem that generates an infringement report and initiates a request for removal.

6. A new media content creation and distribution platform according to claim 1, characterized in that, The user profile construction module (110) includes: A social relationship graph generator (111) that analyzes the user's follow, like, and share relationship chains; An interest prediction model (112) that predicts content preferences based on user dwell time and completion rate; A device feature analyzer (113) that adapts the content format according to the terminal type and network environment.

7. A new media content creation and distribution platform according to claim 1, characterized in that It also includes: A virtual digital human driving module (160) that generates a virtual anchor broadcast video based on user input text; An intelligent voice-over unit (161) that provides more than 50 voice styles and multi-language conversion; An expression and action library (162) that contains more than 2000 pre-set virtual human action templates.

8. A new media content creation and distribution platform according to claim 1, characterized in that, The data analysis center (150) provides: A dissemination path visualization tool (151) that draws the diffusion map of content in the social network; An ROI calculator (152) that calculates the input-output ratio of a single piece of content; An intelligent review report generator (153) that automatically outputs a list of content optimization suggestions.

9. A distribution method of a new media content creation and distribution platform according to any one of claims 1-8, characterized in that, Including the following steps: S1: Extract content keywords and sentiment tendencies through semantic analysis; S2: Match the historical popular content feature library to find similar patterns; S3: Construct a triangular relationship model of platform - user - content using knowledge graph technology; S4: Generate a distribution plan including release time, platform combination, and promotion budget; S5: Dynamically adjust the distribution strategy according to real-time feedback data.