Role characteristic generation method and system

Abstract

The invention discloses a role feature generation method, which comprises the following steps of: selecting an operation mode through an operation mode device, and screening materials; the role analysis device executes a role analysis task on the material to generate multi-version role feature answers; the result processing device performs strip splitting and alignment on the character feature answers; selecting a role characteristic result through the user interaction device for adjustment and customization processing, and generating a role characteristic result of a final version; and the data storage device correspondingly stores the role characteristic result of the final version and the corresponding full-process operation data as role characteristic data. In addition, the invention also discloses a role characteristic generation system. According to the role characteristic generation method, dual modes are utilized to adapt to different use scenes, and efficiency and flexibility are both considered; the whole-process data storage supports history tracing and material reuse, so that the maintenance cost is reduced; and the overall quality and the production efficiency are improved through deep cooperation of artificial intelligence generation and artificial optimization.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method and system for generating character traits. Background Technology

[0002] With the development of technology and the diversification of information access channels, digital media is constantly impacting traditional media, and the public's reading habits are constantly changing, giving rise to audiobooks as a reading format. Audiobooks are electronic publications that use sound as the medium, including audiobooks, audio novels, and so on. Audiobooks overlap with both digital and traditional media, and their unique advantages can meet the needs of various users.

[0003] In existing technologies, the generation of character traits in audiobook production mainly relies on manual writing or simple text extraction. This often results in inefficiency, insufficient accuracy, lack of flexibility, poor traceability, and inadequate human-computer collaboration. When processing characters in batches, manual extraction of the audiobook's illustration content is required, which is time-consuming, labor-intensive, and easily influenced by subjective human factors, leading to a one-sided description of character traits. Existing tools cannot flexibly adjust the scope of analysis materials, making it difficult to balance batch processing and precise customization needs. Furthermore, the lack of a complete workflow record during generation makes subsequent optimization or reuse difficult. Additionally, the lack of an effective mechanism for collaborative work between artificial intelligence (AI) and humans makes it impossible to achieve a balance between efficiency and quality. Summary of the Invention

[0004] Based on this, in order to solve the technical problems in the existing technology, a method for generating character characteristics is proposed, including: The operation mode device provides users with a mode switching entry. Based on the operation commands issued by the user, the corresponding operation mode is selected and the corresponding material processing logic is triggered to filter the materials to be processed. The operation mode device sends the filtered materials and corresponding task commands to the role analysis device and stores the material text list and task commands in the data storage device. The character analysis device performs character analysis on the selected materials to generate multiple versions of character characteristic answers; the character analysis device sends the generated character characteristic answers to the result processing device, and stores the multiple versions of character characteristic answers and large model parameters in the data storage device; The result processing device processes the received role characteristic answers. The result processing includes splitting and aligning, splitting and classifying the core information in each version of the answer and then matching them one by one. The core information includes age, personality and identity. The result processing device sends the role characteristic results obtained after splitting and aligning to the user interaction device, and stores the splitting and alignment operation records and results in the data storage device. User interaction is performed through a user interaction device. Users can select the character characteristics they need to process by making overall selections or segmented selections, and then make adjustments and customizations. The user interaction device outputs the final version of the adjusted and customized character characteristics to the user for confirmation and sends it to the data storage device. The data storage device receives the final version of the character characteristic results and stores them as character characteristic data, corresponding to the full-process operation data.

[0005] In one embodiment, the user selects an operation mode in the operation mode device according to their needs. The operation mode includes a simplified mode or a refined mode. The simplified mode is used for rapid batch processing of character characteristics, while the refined mode is used for precise customization of character characteristics. When the user selects a target character, in the simplified mode, the user directly filters the chapters or paragraphs in the sketchbook and extracts the target character data to perform a character analysis task. In the refined mode, the user customizes the selection of chapters or paragraphs and then extracts the target character data to perform a character analysis task. The materials to be processed include picture books with audiobooks, each containing one or more chapters, and each chapter containing one or more paragraphs; the material processing logic includes: When the user selects and triggers the simplified mode, the system directly filters all chapters or paragraphs related to the target character from the illustration book without requiring any additional user action; the operation mode device will then place the directly filtered materials into a material text list. When a user selects and triggers the fine mode, the system displays a list of all chapters in the picture book. The user then sends task commands through the operation mode device. These task commands include deleting unwanted chapters or paragraphs, selecting a range of chapters or paragraphs, and manually filtering chapters or paragraphs. The operation mode device then places the filtered materials into a material text list.

[0006] In one embodiment, a character analysis task is performed on the screened materials using a character analysis device, specifically including: The character analysis device receives materials sent by the operation mode device and uses the pre-trained large model therein to perform character feature analysis tasks, including: Schedule the analysis tasks to be executed and process them in the order of submission to avoid concurrency conflicts; Extract character attribute information from the material, including age, gender, identity, personality traits, abilities and special skills, and social status; Multiple versions of character characteristic answers are generated based on the extracted character attribute information. Two or more versions of character characteristic answers are generated by adjusting the large model parameters to cover different perspectives. The large model parameters include the character attribute information extraction weights and character description styles.

[0007] In one embodiment, the answer to the character traits in multiple versions is split and aligned, specifically including: The process of breaking down the answers to character characteristics involves segmenting them according to character attributes to ensure that similar information across different versions is on the same level. Character attributes include age, gender, identity, personality traits, abilities and strengths, and social status. Aligning the answers to character traits involves displaying the paragraphs of each version of the answer to the user, allowing the user to visually compare the differences in expression between the different versions of the answer.

[0008] In one embodiment, the user interaction device provides an operation entry point for the user, through which the user adjusts and customizes the role characteristic results generated by the result processing device, specifically including: The system allows users to switch between roles based on their characteristics, enabling them to quickly switch roles via user interaction devices and process multiple roles in batches. The system allows users to select versions of the character traits. Version selection can be done in two ways: overall selection and segment selection. Users can select the complete content of any version directly using the overall selection method, or select and combine segments from different versions using the segment selection method. The user can edit the character trait results, making modifications based on the selected version of the character trait results, including adding or deleting text and adjusting the wording. The results of character characteristics are scored. Users score the results of character characteristics for each version and record the scores. The scores provide data support for the adjustment and iteration of the model. The system provides feedback on the character trait results. If the user is not satisfied with the generated character trait results, they can select to trigger a re-analysis command, which will send the character trait results that need to be re-analyzed back to the character analysis device and re-execute the analysis task.

[0009] In one embodiment, role characteristic data is used for subsequent data traceability and reuse. Role characteristic data includes input data, process data, and output data; wherein, the full-process operation data includes input data and process data. The input data includes a list of text materials selected by the user, including results directly extracted in simplified mode or results manually filtered in fine mode. The process data includes multiple versions of answers, records of splitting and aligning operations, user selection records, scoring data, and modification records; The output data represents the final version of the character traits. In one embodiment, the data storage device feeds back the stored role characteristic data to the operation mode device for retrieval, and the operation mode device reuses the historical data and materials therein based on the role characteristic data it retrieves; the data storage device feeds back the stored role characteristic data to the role analysis device for retrieval, and the role analysis device adjusts the big data model based on the role characteristic data it retrieves.

[0010] In addition, to solve the technical problems in the existing technology, a role characteristic generation system is proposed, including an operation mode device, a role analysis device, a result processing device, a user interaction device, and a data storage device. The operation mode device is connected to the role analysis device, the role analysis device is connected to the result processing device, the result processing device is connected to the user interaction device, and the data storage device is interconnected with the operation mode device, the role analysis device, the result processing device, and the user interaction device. The operation mode device provides users with a mode switching entry. Based on the operation commands issued by the user, the corresponding operation mode is selected and the corresponding material processing logic is triggered to filter the materials to be processed. The operation mode device sends the filtered materials and corresponding task commands to the role analysis device and stores the material text list and task commands in the data storage device. The character analysis device performs character analysis on the selected materials to generate multiple versions of character characteristic answers; the character analysis device sends the generated character characteristic answers to the result processing device, and stores the multiple versions of character characteristic answers and large model parameters in the data storage device; The result processing device processes the received role characteristic answers. The result processing includes splitting and aligning, splitting and classifying the core information in each version of the answer and then matching them one by one. The core information includes age, personality and identity. The result processing device sends the role characteristic results obtained after splitting and aligning to the user interaction device, and stores the splitting and alignment operation records and results in the data storage device. User interaction is performed through a user interaction device. Users can select the character characteristics they need to process by making overall selections or segmented selections, and then make adjustments and customizations. The user interaction device outputs the final version of the adjusted and customized character characteristics to the user for confirmation and sends it to the data storage device. The data storage device receives the final version of the character characteristic results and stores them as character characteristic data, corresponding to the full-process operation data.

[0011] In one embodiment, the user selects an operation mode in the operation mode device according to their needs. The operation mode includes a simplified mode or a refined mode. The simplified mode is used for rapid batch processing of character characteristics, while the refined mode is used for precise customization of character characteristics. When the user selects a target character, in the simplified mode, the user directly filters the chapters or paragraphs in the sketchbook and extracts the target character data to perform a character analysis task. In the refined mode, the user customizes the selection of chapters or paragraphs and then extracts the target character data to perform a character analysis task. The materials to be processed include picture books with audiobooks, each containing one or more chapters, and each chapter containing one or more paragraphs; the material processing logic includes: When the user selects and triggers the simplified mode, the system directly filters all chapters or paragraphs related to the target character from the illustration book without requiring any additional user action; the operation mode device will then place the directly filtered materials into a material text list. When a user selects and triggers the fine mode, the system displays a list of all chapters in the picture book. The user then sends task commands through the operation mode device. These task commands include deleting unwanted chapters or paragraphs, selecting a range of chapters or paragraphs, and manually filtering chapters or paragraphs. The operation mode device then places the filtered materials into a material text list.

[0012] In one embodiment, a character analysis task is performed on the screened materials using a character analysis device, specifically including: The character analysis device receives materials sent by the operation mode device and uses the pre-trained large model therein to perform character feature analysis tasks, including: Schedule the analysis tasks to be executed and process them in the order of submission to avoid concurrency conflicts; Extract character attribute information from the material, including age, gender, identity, personality traits, abilities and special skills, and social status; Multiple versions of character characteristic answers are generated based on the extracted character attribute information. Two or more versions of character characteristic answers are generated by adjusting the large model parameters to cover different perspectives. The large model parameters include the character attribute information extraction weights and character description styles.

[0013] In one embodiment, the answer to the character traits in multiple versions is split and aligned, specifically including: The process of breaking down the answers to character characteristics involves segmenting them according to character attributes to ensure that similar information across different versions is on the same level. Character attributes include age, gender, identity, personality traits, abilities and strengths, and social status. Aligning the answers to character traits involves displaying the paragraphs of each version of the answer to the user, allowing the user to visually compare the differences in expression between the different versions of the answer.

[0014] In one embodiment, the user interaction device provides an operation entry point for the user, through which the user adjusts and customizes the role characteristic results generated by the result processing device, specifically including: The system allows users to switch between roles based on their characteristics, enabling them to quickly switch roles via user interaction devices and process multiple roles in batches. The system allows users to select versions of the character traits. Version selection can be done in two ways: overall selection and segment selection. Users can select the complete content of any version directly using the overall selection method, or select and combine segments from different versions using the segment selection method. The user can edit the character trait results, making modifications based on the selected version of the character trait results, including adding or deleting text and adjusting the wording. The results of character characteristics are scored. Users score the results of character characteristics for each version and record the scores. The scores provide data support for the adjustment and iteration of the model. The user interaction device is connected to the role analysis device; The system provides feedback on the character trait results. If the user is not satisfied with the generated character trait results, they can select to trigger a re-analysis command, which will send the character trait results that need to be re-analyzed back to the character analysis device and re-execute the analysis task.

[0015] In one embodiment, role characteristic data is used for subsequent data traceability and reuse. Role characteristic data includes input data, process data, and output data; wherein, the full-process operation data includes input data and process data. The input data includes a list of text materials selected by the user, including results directly extracted in simplified mode or results manually filtered in fine mode. The process data includes multiple versions of answers, records of splitting and aligning operations, user selection records, scoring data, and modification records; The output data represents the final version of the character traits. In one embodiment, the data storage device feeds back the stored role characteristic data to the operation mode device for retrieval, and the operation mode device reuses the historical data and materials therein based on the role characteristic data it retrieves; the data storage device feeds back the stored role characteristic data to the role analysis device for retrieval, and the role analysis device adjusts the big data model based on the role characteristic data it retrieves.

[0016] Implementing the embodiments of the present invention will have the following beneficial effects: This invention achieves significant technical effects by incorporating dual operation modes, multi-version answer generation and segment alignment, full-process data traceability, and human-computer collaborative interaction mechanisms into character characteristic generation: the simplified mode greatly improves the efficiency of batch character processing, while the refined mode ensures the relevance of results through customized material screening, improving the completeness and accuracy of character characteristic descriptions; the dual modes adapt to different usage scenarios, balancing efficiency and flexibility; the full-process operation data storage supports historical traceability and material reuse, reducing maintenance costs; and the deep collaboration between artificial intelligence generation and human optimization comprehensively improves the overall quality and production efficiency of audiobook character characteristic generation. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] in: Figure 1 This is a flowchart illustrating the character trait generation method in this invention; Figure 2 This is a schematic diagram of the character characteristic generation system in this invention; Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] like Figure 1 As shown, this invention discloses a method for generating character traits, including: The user selects an operation mode in the operation mode device according to their needs. The operation modes include a simplified mode or a refined mode. The simplified mode is used for rapid batch processing of character characteristics, while the refined mode is used for precise customization of character characteristics. The operation mode device provides users with a mode switching entry for simplified mode and refined mode. It selects the corresponding operation mode according to the operation command issued by the user and triggers the material processing logic corresponding to the mode to filter the material to be processed. Specifically, the material to be processed includes picture books with audiobooks, each containing one or more chapters, and each chapter containing one or more paragraphs; For example, when the audiobook is an audio novel, the picture book is usually composed of the original text of the novel, including the dialogue of each character in the original text. Different characters' dialogues are marked with different colors to distinguish the characters, making it easier for the recording staff to record. In general, the picture book is the text material that marks the dialogue of different characters in the novel. The material processing logic includes: When a user selects a target character, in simplified mode, the chapters or paragraphs in the sketchbook are directly filtered and the target character data is extracted for character analysis tasks. In refined mode, the user can customize the selection of chapters or paragraphs and then extract the target character data for character analysis tasks. Specifically, when a user selects and triggers the simplified mode, the system directly filters all chapters or paragraphs related to the target character from the sketchbook without requiring any additional user action; the operation mode device then places the directly filtered materials into a material text list. When the user selects and triggers the fine mode, the system displays a list of all chapters in the picture book. The user sends task instructions through the operation mode device. These task instructions include deleting unnecessary chapters or paragraphs, selecting a range of chapters or paragraphs, and manually filtering chapters or paragraphs. The operation mode device then places the filtered materials into a material text list. After the materials are filtered by the operation mode device, the operation mode device sends the filtered materials and corresponding task instructions to the character analysis device connected to it. The operation mode device stores the material text list and task instructions as the data for the entire operation process in a data storage device connected to it; The character analysis device performs character analysis tasks on the selected materials; the character analysis device schedules the character analysis tasks to be performed and executes them in sequence, and uses the data extracted from the picture book to generate answers to the character characteristics of multiple versions; Specifically, the selected materials are analyzed using a character analysis device, including the following: The character analysis device receives materials sent by the operation mode device and uses the pre-trained large model therein to perform character feature analysis tasks, including: Schedule the analysis tasks to be executed and process them in the order of submission to avoid concurrency conflicts; Extract character attribute information from the material, including age, gender, identity, personality traits, abilities and special skills, and social status; Multiple versions of character characteristic answers are generated based on the extracted character attribute information. Two or more versions of character characteristic answers are generated by adjusting the large model parameters to cover different expression perspectives. The large model parameters include the character attribute information extraction weight and the character description style. The character analysis device sends the generated multiple versions of character characteristic answers to the result processing device connected to it; The character analysis device stores the answers to character characteristics from multiple versions and the parameters of the large model as data for the entire process of data storage in a data storage device connected to it. The result processing device receives the character characteristic answers sent by the analysis device and processes them. The result processing includes splitting and aligning multiple versions of character characteristic answers, breaking down and classifying the core information in each version of the answer, and then matching them one by one. The core information includes age, personality, and identity. Specifically, the answers to the character traits from multiple versions were split and aligned, including: The process of breaking down the answers to character characteristics involves segmenting them according to character attributes to ensure that similar information across different versions is on the same level. Character attributes include age, gender, identity, personality traits, abilities and strengths, and social status. Aligning the answers to character characteristics includes displaying the paragraphs of each version of the answer to the user one by one, so that the user can intuitively compare the differences in expression between the different versions of the answer; The result processing device sends the character characteristic results obtained after striping and alignment to the user interaction device connected to it; The results processing device stores the stripping and alignment operation records and results as the entire process operation data to the data storage device connected to it; User interaction is performed through a user interaction device. Users can select the character characteristics results to be processed by making overall selections or segmented selections, and adjust and customize the content to be processed. User interaction processing enables collaborative operation between AI generation and human optimization; Specifically, the user interaction device provides users with an operation entry point, through which users can adjust and customize the role characteristic results generated by the result processing device, including: The system allows users to switch between roles based on their characteristics, enabling them to quickly switch roles via user interaction devices and process multiple roles in batches. The system allows users to select versions of the character traits. Version selection can be done in two ways: overall selection and segment selection. Users can select the complete content of any version directly using the overall selection method, or select and combine segments from different versions using the segment selection method. The user can edit the character trait results, making modifications based on the selected version of the character trait results, including adding or deleting text and adjusting the wording. The results of character characteristics are scored. Users score the results of character characteristics for each version and record the scores. The scores provide data support for the adjustment and iteration of the model. The system provides feedback on the character trait results. If the user is not satisfied with the generated character trait results, they can select to trigger a re-analysis command, which will send the character trait results that need to be re-analyzed to the analysis device connected to it and re-execute the analysis task. The user interaction device outputs the final version of the character characteristics, after adjustment and customization, to the user for confirmation, and sends it to the data storage device connected to it. The data storage device receives the final version of the character characteristic results output by the user interaction device and stores it as character characteristic data in correspondence with the corresponding full-process operation data; Among them, role characteristic data is used for subsequent data traceability and reuse, and role characteristic data includes input data, process data, and output data; among them, the full-process operation data includes input data and process data; The input data includes a list of text materials selected by the user, including results directly extracted in simplified mode or results manually filtered in fine mode. The process data includes multiple versions of answers, records of splitting and aligning operations, user selection records, scoring data, and modification records; The output data represents the final version of the character traits. Specifically, the data storage device is connected to the operation mode device and the role analysis device, respectively; The data storage device feeds back the stored character characteristic data to the operation mode device for retrieval, and the operation mode device reuses the historical data and materials in the stored character characteristic data according to the character characteristic data it retrieves; The data storage device feeds back the stored role characteristic data to the role analysis device for retrieval, and the role analysis device adjusts the big data model based on the role characteristic data it retrieves.

[0021] In addition, such as Figure 2 As shown, the present invention also discloses a character characteristic generation system, including an operation mode device, a character analysis device, a result processing device, a user interaction device, and a data storage device; wherein, the operation mode device is connected to the character analysis device, the character analysis device is connected to the result processing device, the result processing device is connected to the user interaction device, and the data storage device is interconnected with the operation mode device, the character analysis device, the result processing device, and the user interaction device respectively. The user selects an operation mode in the operation mode device according to their needs. The operation modes include a simplified mode or a refined mode. The simplified mode is used for rapid batch processing of character characteristics, while the refined mode is used for precise customization of character characteristics. The operation mode device provides users with a mode switching entry for simplified mode and refined mode. It selects the corresponding operation mode according to the operation command issued by the user and triggers the material processing logic corresponding to the mode to filter the material to be processed. Specifically, the material to be processed includes picture books with audiobooks, each containing one or more chapters, and each chapter containing one or more paragraphs; The material processing logic includes: When a user selects a target character, in simplified mode, the chapters or paragraphs in the sketchbook are directly filtered and the target character data is extracted for character analysis tasks. In refined mode, the user can customize the selection of chapters or paragraphs and then extract the target character data for character analysis tasks. Specifically, when a user selects and triggers the simplified mode, the system directly filters all chapters or paragraphs related to the target character from the sketchbook without requiring any additional user action; the operation mode device then places the directly filtered materials into a material text list. When the user selects and triggers the fine mode, the system displays a list of all chapters in the picture book. The user sends task instructions through the operation mode device. These task instructions include deleting unnecessary chapters or paragraphs, selecting a range of chapters or paragraphs, and manually filtering chapters or paragraphs. The operation mode device then places the filtered materials into a material text list. After the materials are filtered by the operation mode device, the operation mode device sends the filtered materials and corresponding task instructions to the character analysis device connected to it. The operation mode device stores the material text list and task instructions as the data for the entire operation process in a data storage device connected to it; The character analysis device performs character analysis tasks on the selected materials; the character analysis device schedules the character analysis tasks to be performed and executes them in sequence, and uses the data extracted from the picture book to generate answers to the character characteristics of multiple versions; Specifically, the selected materials are analyzed using a character analysis device, including the following: The character analysis device receives materials sent by the operation mode device and uses the pre-trained large model therein to perform character feature analysis tasks, including: Schedule the analysis tasks to be executed and process them in the order of submission to avoid concurrency conflicts; Extract character attribute information from the material, including age, gender, identity, personality traits, abilities and special skills, and social status; Multiple versions of character characteristic answers are generated based on the extracted character attribute information. Two or more versions of character characteristic answers are generated by adjusting the large model parameters to cover different expression perspectives. The large model parameters include the character attribute information extraction weight and the character description style. The character analysis device sends the generated multiple versions of character characteristic answers to the result processing device connected to it; The character analysis device stores the answers to character characteristics from multiple versions and the parameters of the large model as data for the entire process of data storage in a data storage device connected to it. The result processing device receives the character characteristic answers sent by the analysis device and processes them. The result processing includes splitting and aligning multiple versions of character characteristic answers, breaking down and classifying the core information in each version of the answer, and then matching them one by one. The core information includes age, personality, and identity. Specifically, the answers to the character traits from multiple versions were split and aligned, including: The process of breaking down the answers to character characteristics involves segmenting them according to character attributes to ensure that similar information across different versions is on the same level. Character attributes include age, gender, identity, personality traits, abilities and strengths, and social status. Aligning the answers to character characteristics includes displaying the paragraphs of each version of the answer to the user one by one, so that the user can intuitively compare the differences in expression between the different versions of the answer; The result processing device sends the character characteristic results obtained after striping and alignment to the user interaction device connected to it; The results processing device stores the stripping and alignment operation records and results as the entire process operation data to the data storage device connected to it; User interaction is performed through a user interaction device. Users can select the character characteristics results to be processed by making overall selections or segmented selections, and adjust and customize the content to be processed. User interaction processing enables collaborative operation between AI generation and human optimization; Specifically, the user interaction device provides users with an operation entry point, through which users can adjust and customize the role characteristic results generated by the result processing device, including: The system allows users to switch between roles based on their characteristics, enabling them to quickly switch roles via user interaction devices and process multiple roles in batches. The system allows users to select versions of the character traits. Version selection can be done in two ways: overall selection and segment selection. Users can select the complete content of any version directly using the overall selection method, or select and combine segments from different versions using the segment selection method. The user can edit the character trait results, making modifications based on the selected version of the character trait results, including adding or deleting text and adjusting the wording. The user interaction device is connected to the role analysis device; The results of character characteristics are scored. Users score the results of character characteristics in each version and record the scores. The scores provide data support for the adjustment and iteration of the model in the character analysis device. The system provides feedback on the character trait results. If the user is not satisfied with the generated character trait results, they can select to trigger a re-analysis command, which will send the character trait results that need to be re-analyzed to the connected character analysis device and re-execute the analysis task. The user interaction device outputs the final version of the character characteristics, after adjustment and customization, to the user for confirmation, and sends it to the data storage device connected to it. The data storage device receives the final version of the character characteristic results output by the user interaction device and stores it as character characteristic data in correspondence with the corresponding full-process operation data; Among them, role characteristic data is used for subsequent data traceability and reuse, and role characteristic data includes input data, process data, and output data; among them, the full-process operation data includes input data and process data; The input data includes a list of text materials selected by the user, including results directly extracted in simplified mode or results manually filtered in fine mode. The process data includes multiple versions of answers, records of splitting and aligning operations, user selection records, scoring data, and modification records; The output data represents the final version of the character traits. Specifically, the data storage device is connected to the operation mode device and the role analysis device, respectively; The data storage device feeds back the stored character characteristic data to the operation mode device for retrieval, and the operation mode device reuses the historical data and materials in the stored character characteristic data according to the character characteristic data it retrieves; The data storage device feeds back the stored role characteristic data to the role analysis device for retrieval, and the role analysis device adjusts the big data model based on the role characteristic data it retrieves.

[0022] This invention is an AI-driven character trait analysis system. It combines a simplified and refined dual operation mode, multi-version generation and segment alignment, and full-process data recording and traceability for generating audiobook character traits. Through the collaboration of artificial intelligence (AI) and human expertise, it achieves efficient, accurate, and flexible character trait generation.

[0023] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating character traits, characterized in that, include: The operation mode device provides users with a mode switching entry. Based on the operation commands issued by the user, the corresponding operation mode is selected and the corresponding material processing logic is triggered to filter the materials to be processed. The operation mode device sends the filtered materials and corresponding task commands to the role analysis device and stores the material text list and task commands in the data storage device. The character analysis device performs a character analysis task on the selected materials to generate multiple versions of character characteristic answers; the character analysis device sends the generated character characteristic answers to the result processing device, and stores the multiple versions of character characteristic answers and large model parameters in the data storage device; The result processing device processes the received role characteristic answers. The result processing includes splitting and aligning, splitting and classifying the core information in each version of the answer and then matching them one by one. The core information includes age, personality and identity. The result processing device sends the role characteristic results obtained after splitting and aligning to the user interaction device, and stores the splitting and alignment operation records and results in the data storage device. User interaction is performed through a user interaction device. Users can select the character characteristics they need to process by making overall selections or segmented selections, and then make adjustments and customizations. The user interaction device outputs the final version of the adjusted and customized character characteristics to the user for confirmation and sends it to the data storage device. The data storage device receives the final version of the character characteristic results and stores them as character characteristic data, corresponding to the full-process operation data.

2. The character trait generation method according to claim 1, Its features are, The user can select an operation mode in the operation mode device according to their needs. The operation modes include a simplified mode and a refined mode. The simplified mode is used for quick batch processing of character characteristics, while the refined mode is used for precise customization of character characteristics. When the user selects a target character, in the simplified mode, the user can directly filter the chapters or paragraphs in the picture book and extract the target character data for the character analysis task. In the refined mode, the user can customize the selection of chapters or paragraphs and extract the target character data for the character analysis task. The materials to be processed include picture books with audiobooks, each containing one or more chapters, and each chapter containing one or more paragraphs; the material processing logic includes: When the user selects and triggers the simplified mode, the system directly filters all chapters or paragraphs related to the target character from the illustration book without requiring any additional user action; the operation mode device will then place the directly filtered materials into a material text list. When a user selects and triggers the fine mode, the system displays a list of all chapters in the picture book. The user then sends task commands through the operation mode device. These task commands include deleting unwanted chapters or paragraphs, selecting a range of chapters or paragraphs, and manually filtering chapters or paragraphs. The operation mode device then places the filtered materials into a material text list.

3. The character trait generation method according to claim 1, characterized in that, The selected materials are analyzed using a character analysis device, specifically including: The character analysis device receives materials sent by the operation mode device and uses the pre-trained large model therein to perform character feature analysis tasks, including: Schedule the analysis tasks to be executed and process them in the order of submission to avoid concurrency conflicts; Extract character attribute information from the material, including age, gender, identity, personality traits, abilities and special skills, and social status; Multiple versions of character characteristic answers are generated based on the extracted character attribute information. Two or more versions of character characteristic answers are generated by adjusting the large model parameters to cover different perspectives. The large model parameters include the character attribute information extraction weights and character description styles.

4. The character trait generation method according to claim 1, characterized in that, The answers to the character traits from multiple versions were broken down and aligned, specifically... include: The process of breaking down the answers to character characteristics involves segmenting them according to character attributes to ensure that similar information across different versions is on the same level. Character attributes include age, gender, identity, personality traits, abilities and strengths, and social status. Aligning the answers to character traits involves displaying the paragraphs of each version of the answer to the user, allowing the user to visually compare the differences in expression between the different versions of the answer.

5. The character trait generation method according to claim 1, characterized in that, The user interaction device provides users with an entry point for operation. Through this entry point, users can adjust and customize the role characteristic results generated by the result processing device, specifically including: The system allows users to switch between roles based on their characteristics, enabling them to quickly switch roles via user interaction devices and process multiple roles in batches. The system allows users to select versions of the character traits. Version selection can be done in two ways: overall selection and segment selection. Users can select the complete content of any version directly using the overall selection method, or select and combine segments from different versions using the segment selection method. The user can edit the character trait results, making modifications based on the selected version of the character trait results, including adding or deleting text and adjusting the wording. The results of character characteristics are scored. Users score the results of character characteristics for each version and record the scores. The scores provide data support for the adjustment and iteration of the model. The system provides feedback on the character trait results. If the user is not satisfied with the generated character trait results, they can select to trigger a re-analysis command, which will send the character trait results that need to be re-analyzed back to the character analysis device and re-execute the analysis task.

6. A character trait generation system, characterized in that, It includes an operation mode device, a role analysis device, a result processing device, a user interaction device, and a data storage device; The operation mode device is connected to the role analysis device, the role analysis device is connected to the result processing device, the result processing device is connected to the user interaction device, and the data storage device is interconnected with the operation mode device, the role analysis device, the result processing device, and the user interaction device. The operation mode device provides users with a mode switching entry. Based on the operation commands issued by the user, the corresponding operation mode is selected and the corresponding material processing logic is triggered to filter the materials to be processed. The operation mode device sends the filtered materials and corresponding task commands to the role analysis device and stores the material text list and task commands in the data storage device. The character analysis device performs a character analysis task on the selected materials to generate multiple versions of character characteristic answers; the character analysis device sends the generated character characteristic answers to the result processing device, and stores the multiple versions of character characteristic answers and large model parameters in the data storage device; The result processing device processes the received role characteristic answers. The result processing includes splitting and aligning, splitting and classifying the core information in each version of the answer and then matching them one by one. The core information includes age, personality and identity. The result processing device sends the role characteristic results obtained after splitting and aligning to the user interaction device, and stores the splitting and alignment operation records and results in the data storage device. User interaction is performed through a user interaction device. Users can select the character characteristics they need to process by making overall selections or segmented selections, and then make adjustments and customizations. The user interaction device outputs the final version of the adjusted and customized character characteristics to the user for confirmation and sends it to the data storage device. The data storage device receives the final version of the character characteristic results and stores them as character characteristic data, corresponding to the full-process operation data.

7. The character trait generation system according to claim 6, Its features are, in, The user selects an operation mode in the operation mode device according to their needs. The operation modes include a simplified mode or a refined mode. The simplified mode is used for quick batch processing of character characteristics, while the refined mode is used for precise customization of character characteristics. When the user selects a target character, in the simplified mode, the user can directly filter the chapters or paragraphs in the sketchbook and extract the target character data to perform the character analysis task. In the refined mode, the user can customize the selection of chapters or paragraphs and extract the target character data to perform the character analysis task. The materials to be processed include picture books with audiobooks, each containing one or more chapters, and each chapter containing one or more paragraphs; the material processing logic includes: When the user selects and triggers the simplified mode, the system directly filters all chapters or paragraphs related to the target character from the illustration book without requiring any additional user action; the operation mode device will then place the directly filtered materials into a material text list. When a user selects and triggers the fine mode, the system displays a list of all chapters in the picture book. The user then sends task commands through the operation mode device. These task commands include deleting unwanted chapters or paragraphs, selecting a range of chapters or paragraphs, and manually filtering chapters or paragraphs. The operation mode device then places the filtered materials into a material text list.

8. The character trait generation system according to claim 6, characterized in that, The selected materials are analyzed using a character analysis device, specifically including: The character analysis device receives materials sent by the operation mode device and uses the pre-trained large model therein to perform character feature analysis tasks, including: Schedule the analysis tasks to be executed and process them in the order of submission to avoid concurrency conflicts; Extract character attribute information from the material, including age, gender, identity, personality traits, abilities and special skills, and social status; Multiple versions of character characteristic answers are generated based on the extracted character attribute information. Two or more versions of character characteristic answers are generated by adjusting the large model parameters to cover different perspectives. The large model parameters include the character attribute information extraction weights and character description styles.

9. The character trait generation system according to claim 6, characterized in that, The answers to the character traits from multiple versions were broken down and aligned, specifically... include: The process of breaking down the answers to character characteristics involves segmenting them according to character attributes to ensure that similar information across different versions is on the same level. Character attributes include age, gender, identity, personality traits, abilities and strengths, and social status. Aligning the answers to character traits involves displaying the paragraphs of each version of the answer to the user, allowing the user to visually compare the differences in expression between the different versions of the answer.

10. The character characteristic generation system according to claim 6, characterized in that, The user interaction device provides users with an entry point for operation. Through this entry point, users can adjust and customize the role characteristic results generated by the result processing device, specifically including: The system allows users to switch between roles based on their characteristics, enabling them to quickly switch roles via user interaction devices and process multiple roles in batches. The system allows users to select versions of the character traits. Version selection can be done in two ways: overall selection and segment selection. Users can select the complete content of any version directly using the overall selection method, or select and combine segments from different versions using the segment selection method. The user can edit the character trait results, making modifications based on the selected version of the character trait results, including adding or deleting text and adjusting the wording. The results of character characteristics are scored. Users score the results of character characteristics for each version and record the scores. The scores provide data support for the adjustment and iteration of the model. The user interaction device is connected to the role analysis device; The system provides feedback on the character trait results. If the user is not satisfied with the generated character trait results, they can select to trigger a re-analysis command, which will send the character trait results that need to be re-analyzed back to the character analysis device and re-execute the analysis task.

Images