CN117610535B - AI writing method and system based on extensible model - Google Patents

Abstract

The invention provides an AI writing method and system based on an expandable model, which are used for receiving a customization request of a demand end, acquiring block configuration information of the demand end based on the customization request, and updating an initial template to obtain the expandable template; obtaining the customization demand information of the demand end on the corresponding customization block in the expandable template, wherein the customization demand information comprises necessary customization information and unnecessary customization information, and updating the expandable template according to the customization demand information to obtain the customization template; receiving a writing request of a demand end, responding to the writing request, acquiring customized content data of a corresponding customized block based on customized demand information of a customized template, and updating the customized template according to the customized content data to obtain customized file data; and calling an automatic calibration model to calibrate the customized content data in the customized file data to obtain calibration file data, and sending the calibration file data to the demand end.

Description

AI writing method and system based on extensible model

Technical Field

The invention relates to an artificial intelligence technology, in particular to an AI writing method and system based on an extensible model.

Background

With the rapid development of artificial intelligence technology, AI writing has become a popular research field, and the writing efficiency of users can be greatly improved through AI writing.

However, manual intervention in the AI writing process is rare, and after the contribution data is completed, the user may also spend a lot of time adjusting the contribution data, for example, the writing requirements and writing formats of the user may be different for different parts of the article, and the user may need to use the AI writing for multiple times according to different writing requirements and writing formats to complete the final article data, so that the user cannot perform customized adjustment on the AI writing.

Therefore, how to customize and adjust the AI writing according to the user requirements, so that the user can combine with the adjustment strategy of the user to conduct the generation guidance of writing when the user performs the AI writing, and the problem to be solved is urgent.

Disclosure of Invention

The embodiment of the invention provides an AI writing method and system based on an extensible model, which can be used for carrying out customized adjustment on AI writing in combination with user requirements, so that a user can carry out generation guidance of writing in combination with an adjustment strategy of the user when carrying out AI writing.

In a first aspect of the embodiment of the present invention, an AI writing method based on a scalable model is provided, including:

Receiving a customization request of a demand end, acquiring block configuration information of the demand end based on the customization request, and updating an initial template to obtain an expandable template;

Obtaining the customization demand information of the demand end for the corresponding customization block in the expandable template, wherein the customization demand information comprises necessary customization information and unnecessary customization information, and updating the expandable template according to the customization demand information to obtain the customization template;

receiving a writing request of the demand end, responding to the writing request, acquiring customized content data of a corresponding customized block based on customized demand information of the customized template, and updating the customized template according to the customized content data to obtain customized file data;

And retrieving an automatic calibration model to calibrate the customized content data in the customized file data to obtain calibration file data, and sending the calibration file data to the demand end.

Optionally, in one possible implementation manner of the first aspect, receiving a customization request of a demand end, acquiring block configuration information of the demand end based on the customization request, and updating an initial template to obtain an expandable template, including:

receiving a customization request of the demand end, calling the initial template and sending the initial template to the demand end;

acquiring specification information input by the demand end to a preset rectangular frame, and updating the specification of the preset rectangular frame according to the specification information to obtain a customized rectangular frame;

Receiving positioning information of the demand end on the customized rectangular frame in the initial template, filling the customized rectangular frame to a corresponding position in the initial template according to the positioning information, and taking a region corresponding to the customized rectangular frame as a customized block;

And obtaining block configuration information according to the customized blocks, and updating the initial template based on the block configuration information to obtain the expandable template.

Optionally, in one possible implementation manner of the first aspect, obtaining the custom requirement information of the requirement end on the corresponding custom block in the expandable template, where the custom requirement information includes necessary custom information and unnecessary custom information, and updating the expandable template according to the custom requirement information to obtain the custom template includes:

Receiving a demand customization request of the demand end for a corresponding customization block in the expandable template, calling a necessary demand column and a non-necessary demand column corresponding to the corresponding customization block, and sending the necessary demand column and the non-necessary demand column to the demand end;

acquiring necessary customization information input by the demand end based on the necessary demand column and unnecessary customization information input by the unnecessary demand column, and acquiring customization demand information according to the necessary customization information and the unnecessary customization information;

and updating the expandable template according to the customization demand information to obtain a customization template.

Optionally, in one possible implementation manner of the first aspect, acquiring the necessary customization information input by the requirement end based on the necessary requirement column and the unnecessary customization information input by the unnecessary requirement column includes:

The method comprises the steps of obtaining requirement information input by a requirement end to all necessary setting data in a necessary requirement column to obtain necessary customization information, wherein the necessary customization information at least comprises a customization writing direction;

and determining unnecessary setting data selected by the demand end in the unnecessary demand column as target setting data, and acquiring demand information input by the demand end on the target setting data to obtain unnecessary customized information.

Optionally, in one possible implementation manner of the first aspect, receiving a writing request of the requirement end, responding to the writing request, and obtaining custom content data of a corresponding custom block based on custom requirement information of the custom template, and updating the custom template according to the custom content data to obtain custom file data, including:

Receiving a writing request of the demand end, and arranging each customized block in the customized template according to a preset arrangement direction to obtain a customized block sequence;

According to the arrangement sequence of each customized block in the customized block sequence, arranging the customized demand information corresponding to each customized block to obtain a customized demand sequence;

Generating demand content data according to each piece of customization demand information, and arranging the demand content data corresponding to each piece of customization demand information based on the customization demand sequence to obtain a demand content sequence;

Sequentially filling each piece of required content data in the required content sequence into corresponding customized blocks in the customized block sequence to obtain customized content data corresponding to the corresponding customized blocks, and updating the format of the customized content data according to a customized format corresponding to the corresponding customized required information;

and updating the customization template according to the customization content data to obtain customization file data.

Optionally, in one possible implementation manner of the first aspect, receiving the writing request of the demand end, and arranging each custom block in the custom template according to a preset arrangement direction to obtain a custom block sequence, including:

Acquiring block center points corresponding to the customized blocks in the customized template;

And arranging the customized blocks according to the arrangement positions of the block center points in the preset arrangement direction to obtain a customized block sequence.

Optionally, in one possible implementation manner of the first aspect, retrieving an automatic calibration model to calibrate custom content data in the custom file data to obtain calibration file data, and sending the calibration file data to the demand end includes:

when the automatic calibration model is called and the custom block fails to meet the filling requirement of the custom content data, custom calibration information is generated;

responding to the customized calibration information, and acquiring a customized format corresponding to customized demand information of the corresponding customized content data;

If the customized format contains font customization information, performing block adjustment on the customized block to obtain an adjustment block, and filling the customized content data into the adjustment block to obtain first calibration data;

If the customized format does not have the font customization information, performing font adjustment on the customized content data to obtain calibration content data, and filling the calibration content data into the customized block to obtain second calibration data;

and carrying out custom data calibration on the corresponding custom content data according to the first calibration data or the second calibration data to obtain calibration file data, and sending the calibration file data to the demand end.

Optionally, in one possible implementation manner of the first aspect, if font customization information exists in the customization format, performing block adjustment on the customization block to obtain an adjustment block, and filling the customization content data into the adjustment block to obtain first calibration data, including:

If the customized format contains font customization information, acquiring the number of content words which cannot be filled into the customized block in the customized content data;

Obtaining a distance coefficient according to the ratio of the content word number to the reference word number, and obtaining a delay distance based on the distance coefficient and the reference distance;

Carrying out extension processing on the frame wires positioned in the preset arrangement direction in the customized block according to the preset arrangement direction and the extension distance to obtain extension frame wires, and updating the customized rectangular frame corresponding to the customized block according to the extension frame wires to obtain an adjusted rectangular frame;

And determining an area corresponding to the adjustment rectangular frame as an adjustment block, and filling the customized content data into the adjustment block to obtain first calibration data.

Optionally, in one possible implementation manner of the first aspect, if font customization information does not exist in the customization format, performing font adjustment on the customization content data to obtain calibration content data, and filling the calibration content data into the customization block to obtain second calibration data, including:

If the customized format does not contain the font customized information, a preset reduced list is called, wherein the preset reduced list comprises a plurality of preset reduced scales;

Sequentially acquiring corresponding preset reduction ratios from large to small as target reduction ratios, and reducing and adjusting fonts of the customized content data to obtain reduced content data;

and stopping font reduction processing of the reduced content data when the customized block meets the filling requirement of the reduced content data, and filling the reduced content data into the customized block to obtain second calibration data.

In a second aspect of the embodiment of the present invention, an AI authoring system based on a scalable model is provided, including:

the configuration module is used for receiving a customization request of a demand end, acquiring block configuration information of the demand end based on the customization request, and updating an initial template to obtain an expandable template;

the demand module is used for acquiring the customization demand information of the demand end for the corresponding customization block in the expandable template, wherein the customization demand information comprises necessary customization information and unnecessary customization information, and the expandable template is updated according to the customization demand information to obtain the customization template;

the customization module is used for receiving a writing request of the demand end, responding to the writing request, acquiring customized content data of a corresponding customized block based on customization demand information of the customization template, and updating the customization template according to the customized content data to obtain customized file data;

and the calibration module is used for retrieving an automatic calibration model to perform custom data calibration on custom content data in the custom file data to obtain calibration file data, and transmitting the calibration file data to the demand end.

The beneficial effects of the invention are as follows:

1. The invention can carry out customized adjustment on the AI writing according to the user requirement, so that the user can carry out the generation guidance of the writing according to the adjustment strategy of the user when carrying out the AI writing. Specifically, the invention can generate the corresponding custom template in combination with the writing requirement of the user, and the custom template is used for generating the writing guidance, so that the corresponding writing data can be generated in combination with the position requirement and the content requirement of each custom block in the custom template by the user, the writing guidance can be performed according to the adjustment strategy of the user when writing next time, the training of the writing model is realized, and the writing efficiency of the user is improved. And the invention can calibrate the customized content data of the customized file data, so that the obtained file data can be distributed more regularly, and the user can conveniently check the written file data.

2. When the custom template is generated, a plurality of custom blocks are generated by combining the specification requirement and the position requirement of the user, and corresponding custom requirement information is input for each custom region according to the content requirement of the user, so that the obtained custom template can be attached to the writing requirement of the user, and the writing generation instruction can be performed according to the custom template. When a user inputs customization demand information, necessary customization information and unnecessary customization information can be input in combination with the self demand, corresponding customization content data can be generated subsequently through the necessary customization information and the unnecessary customization information, and the corresponding customization content data is filled into the corresponding customization block to obtain writing file data, so that the writing efficiency of the user can be improved.

3. When the customized file data is calibrated, the region size of the corresponding customized region or the font size of the customized content data is adjusted when the number of the filling words of the customized region is judged to be incapable of meeting the number of the filling words of the customized content data, so that the adjusted customized region can meet the filling requirements of the customized content data, and finally the obtained file data can be arranged regularly, thereby facilitating the user to check the writing data.

Drawings

FIG. 1 is a schematic flow chart of an AI writing method based on a scalable model provided by an embodiment of the invention;

FIG. 2 is a schematic diagram of a scalable template provided by an embodiment of the present invention;

FIG. 3 is a schematic diagram of an AI writing system based on a scalable model according to an embodiment of the invention

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Referring to fig. 1, a flowchart of an AI writing method based on a scalable model according to an embodiment of the present application is shown, where an execution subject of the method shown in fig. 1 may be a software and/or hardware device. The execution body of the present application may include, but is not limited to, at least one of: user equipment, network equipment, etc. The user device may include, but is not limited to, a computer, a smart phone, a Personal Digital Assistant (PDA) DIGITAL ASSISTANT, and the above-mentioned electronic device. The network device may include, but is not limited to, a single network server, a server group of multiple network servers, or a cloud of a large number of computers or network servers based on cloud computing, where cloud computing is one of distributed computing, and a super virtual computer consisting of a group of loosely coupled computers. This embodiment is not limited thereto. The method comprises the steps S1 to S4, and specifically comprises the following steps:

s1, receiving a customization request of a demand end, acquiring block configuration information of the demand end based on the customization request, and updating an initial template to obtain an expandable template.

In order to generate a guide for writing by combining with an adjustment strategy of a user when the user performs AI writing, the method can generate a corresponding custom template by combining with the writing requirement of the user, and then generate a guide for writing the AI writing according to the custom template, so that the generated writing file can be attached to the requirement of the user, and the writing efficiency of the user is improved.

Specifically, when the customized template is generated, the initial template can be updated according to the block configuration information input by the user based on the requirement end to obtain the expandable template, and then the expandable template is further subjected to data customization by combining the writing requirement of the user to obtain the customized template.

In practical application, the demand end may be intelligent equipment such as a computer, and the block configuration information refers to block configuration information of a writing area in an initial template by a user.

The specific implementation manner of step S1 based on the above embodiment may be:

S11, receiving a customization request of the demand end, calling the initial template and sending the initial template to the demand end.

In practical application, the initial template may be a preset template, and the user may determine the block data to be written in the initial template, and fill the corresponding block with the writing content to generate the corresponding file data.

S12, acquiring specification information input by the demand end to a preset rectangular frame, and updating the specification of the preset rectangular frame according to the specification information to obtain a customized rectangular frame.

It can be understood that the writing data of each block may be different when the user performs writing, so that when the block is obtained, the preset rectangular frame can be adjusted according to the specification requirement of the user to obtain a customized rectangular frame, and a corresponding customized block is obtained according to the customized rectangular frame later.

In practical application, the specification information may include the length and width specifications of the rectangular frame, and the user may perform corresponding setting in combination with the own requirements.

S13, receiving positioning information of the demand end on the customized rectangular frame in the initial template, filling the customized rectangular frame to a corresponding position in the initial template according to the positioning information, and taking an area corresponding to the customized rectangular frame as a customized block.

It can be understood that a user may set a plurality of blocks to customize the writing data, so after the custom rectangular frame is obtained, the custom rectangular frame can be positioned to a corresponding position in the initial template in combination with positioning information of the user, and then an area corresponding to the custom rectangular frame is used as a custom block, and the custom block is used for customizing the writing data.

In some embodiments, the positioning information may be a positioning point selected by the user for the customized rectangular frame in the initial template, then positioning a center point of the customized rectangular frame through the positioning point, filling the customized rectangular frame into the initial template, and taking an area corresponding to the customized rectangular frame as the customized block.

S14, obtaining block configuration information according to the customized blocks, and updating the initial template based on the block configuration information to obtain the expandable template.

Referring to fig. 2, a schematic diagram of a scalable template is provided in an embodiment of the present invention. As can be seen from fig. 2, a user can customize a plurality of corresponding customized rectangular frames, and correspondingly position each customized rectangular frame in combination with its own positioning requirement, so that specification information and position information of the customized block can be correspondingly customized, for example, customized block 1, customized block 2 and customized block 3 in fig. 2.

By the mode, the block data of the custom writing data can be correspondingly customized according to the requirements of users.

S2, obtaining the customization demand information of the demand end for the corresponding customization block in the expandable template, wherein the customization demand information comprises necessary customization information and unnecessary customization information, and updating the expandable template according to the customization demand information to obtain the customization template.

After the expandable template is obtained, the corresponding custom block can be updated by combining the requirement data of the custom block by a user, so that the generated writing file can be attached to the writing requirement of the user by generating guidance of writing through the updated custom template.

The custom requirement information refers to requirement data input by a user to the custom block, and the requirement data may include requirement data such as writing direction, writing format, and the like. It will be appreciated that, since some of the required data may have to be entered by the user when customizing the authoring data, and some of the required data may not have to be entered, the customization requirement information may be divided into necessary customization information and unnecessary customization information, so that the expandable template may be further updated by the necessary customization information and the unnecessary customization information to obtain the customization template.

In some embodiments, the custom templates described above may be obtained by:

S21, receiving a demand customization request of the demand end for a corresponding customization block in the expandable template, calling a necessary demand column and a non-necessary demand column corresponding to the customization block, and sending the necessary demand column and the non-necessary demand column to the demand end.

It can be understood that the user may input different requirement information to different custom blocks, so after receiving a requirement customization request of the user for a corresponding custom block, the user may invoke a necessary requirement column and a non-necessary requirement column corresponding to the corresponding custom block and send the necessary requirement column to the user to input the requirement information.

The necessary requirement column comprises requirement information which is required to be input when writing customization is carried out, and the unnecessary requirement column comprises requirement information which is not required to be input when writing customization is carried out.

In practical application, the necessary requirement column and the unnecessary requirement column may be preset, the necessary requirement column and the unnecessary requirement column may include a plurality of preset setting data, and then the user inputs corresponding requirement information to the corresponding setting data, and the scheme for the necessary requirement column and the unnecessary requirement column will be described in more detail in the following scheme.

S22, acquiring necessary customization information input by the demand end based on the necessary demand column and unnecessary customization information input by the unnecessary demand column, and acquiring customization demand information according to the necessary customization information and the unnecessary customization information.

It is worth mentioning that the user has to input the custom information in the necessary requirement column, but the user can selectively input the custom information in the unnecessary requirement column. The necessary customization information refers to requirement information input by a user based on a necessary requirement column, and the unnecessary customization information refers to requirement information input by the user based on an unnecessary requirement column.

The specific implementation manner of "obtaining the necessary customization information input by the requirement end based on the necessary requirement column and the unnecessary customization information input by the unnecessary requirement column" in step S22 based on the above embodiment may be:

s221, obtaining the requirement information of the requirement end for inputting the necessary setting data in the necessary requirement column to obtain necessary customization information, wherein the necessary customization information at least comprises a customization writing direction.

In practical applications, the necessary setting data may be preset. It can be understood that when the user performs the writing customization, the user cannot customize the writing content without the writing direction, so that the necessary setting data at least needs to include the writing direction setting data, and then the user obtains the customized writing direction through the requirement information input by the user on the writing direction setting data.

For example, the user's demand information for writing direction setting data input may be to customize title data on a certain aspect, and then the generation customization of the corresponding title may be performed later through the demand information input by the user.

In some embodiments, when the necessary customization information is obtained, the corresponding input interface may be called by the user to click information of the necessary setting data in the necessary requirement column, and sent to the requirement end, and then the necessary customization information corresponding to the corresponding necessary setting data is obtained through the requirement information input in the input interface by the requirement end.

S222, determining unnecessary setting data selected by the demand end in the unnecessary demand column as target setting data, and acquiring demand information input by the demand end on the target setting data to obtain unnecessary customization information.

In practical applications, the unnecessary setting data may also be preset, which may include format setting data, word count setting data, and the like. It is understood that the setting data such as format and word number are not necessarily set, and the writing content can be customized without the setting data, so that when the user inputs the unnecessary customization information, the user can select the corresponding unnecessary setting data to input the corresponding requirement information according to the requirement of the user.

For example, the user may select the format setting data in the unnecessary requirement column, and then input the requirement information of the format data accordingly, for example, the user may set the format information of the font format, the paragraph format, the font style, and the like accordingly.

In some embodiments, when obtaining the unnecessary customized information, the user may also call the corresponding input interface to the requirement end through clicking information of the unnecessary setting data in the unnecessary requirement column, and then obtain the unnecessary customized information corresponding to the corresponding unnecessary setting data through the requirement information input in the input interface by the requirement end.

S23, updating the expandable template according to the customization demand information to obtain a customization template.

The customized template obtained through the method can be combined with the writing requirements of the user to correspondingly customize the content information and the position information of each writing data, so that the obtained writing file can be attached to the writing requirements of the user, and the writing efficiency of the user is improved.

S3, receiving a writing request of the demand end, responding to the writing request, acquiring customized content data of a corresponding customized block based on customized demand information of the customized template, and updating the customized template according to the customized content data to obtain customized file data.

The customized content data refers to content data generated according to the customized demand information of the corresponding customized block in the customized template. In practical applications, the customized content data may be text content data. When the custom template is updated, content data generated by corresponding custom demand information can be filled into the custom template to obtain custom file data.

In some embodiments, the custom file data described above may be obtained by:

S31, receiving a writing request of the demand end, and arranging all the custom blocks in the custom template according to a preset arrangement direction to obtain a custom block sequence.

In practical applications, the preset arrangement direction may be a top-to-bottom direction. The method comprises the steps of arranging each customized block according to a preset arrangement direction to obtain a customized block sequence, wherein content data generated by corresponding customized demand information can be filled into the corresponding customized blocks in the follow-up process.

The specific implementation manner of step S31 based on the above embodiment may be:

s311, obtaining block center points corresponding to the customized blocks in the customized template.

In some embodiments, the block center point may be obtained by customizing the intersection of block diagonals.

S312, arranging the customized blocks according to the arrangement positions of the center points of the blocks in the preset arrangement direction to obtain a customized block sequence.

For example, when the preset arrangement direction is from top to bottom, the custom block corresponding to the center point of the first block arranged in the preset arrangement direction is arranged at the first position, the custom block corresponding to the center point of the second block arranged in the preset arrangement direction is arranged at the second position, and so on to obtain the custom block sequence.

In some embodiments, if there are a plurality of blocks with the same arrangement positions of center points, the corresponding custom blocks may be arranged randomly.

S32, according to the arrangement sequence of each customized block in the customized block sequence, arranging the customized demand information corresponding to each customized block to obtain a customized demand sequence.

It can be appreciated that, in order to enable content data generated according to the customization requirement information to be filled into the corresponding customization block, each customization requirement information may be arranged according to the customization block sequence to obtain the customization requirement sequence, so that content data generated by the customization requirement sequence may be filled into the corresponding customization block.

S33, generating demand content data according to the customization demand information, and arranging the demand content data corresponding to the customization demand information based on the customization demand sequence to obtain a demand content sequence.

The required content data generated by each piece of customized required information may be different, for example, when the required information such as the customized writing direction and the customized format data corresponding to the customized required information are different, the generated required content data may also be different, so that in order to fill the required content data into the corresponding customized block, the required content data may be arranged according to the customized required sequence to obtain the required content sequence, so that the required content sequence and the customized block sequence are mutually corresponding.

And S34, sequentially filling each piece of required content data in the required content sequence into the corresponding customized block in the customized block sequence to obtain customized content data corresponding to the corresponding customized block, and carrying out format updating on the customized content data according to a customized format corresponding to the corresponding customized required information.

When each demand content data in the demand content sequence is sequentially filled into the corresponding custom blocks in the custom block sequence, the demand content data and the custom blocks with the same arrangement positions can be corresponding, and then the corresponding demand content data is filled into the corresponding custom blocks.

It will be appreciated that since it is also possible for the user to customize the format of the customized content data, the customized content data may also be updated in format by the customized format.

And S35, updating the customization template according to the customization content data to obtain customization file data.

By the method, AI writing can be quickly generated in combination with the writing requirements of the user, so that the generated writing data can be attached to the requirements of the user, and a customized writing model related to the user can be generated according to the adjustment of the user, so that the writing can be guided according to the adjustment strategy of the user in the next writing, and training of the writing model is realized.

S4, retrieving an automatic calibration model to calibrate the customized content data in the customized file data to obtain calibration file data, and sending the calibration file data to the demand end.

The automatic calibration model is a model for carrying out data calibration on customized content data in customized file data, and can calibrate the customized content data which does not meet filling requirements, so that finally obtained calibrated file data can be arranged regularly, and the writing requirements of users are met.

Specifically, in some embodiments, step S4 may be implemented through steps S41 to S45, which is specifically as follows:

s41, when the automatic calibration model is called to judge that the custom block fails to meet the filling requirement of the custom content data, custom calibration information is generated.

The above-mentioned determination that the custom block fails to meet the filling requirement of the custom content data refers to a case where the number of filling words of the custom block cannot meet the number of filling words of the custom content data, that is, a case where all text data of the custom content data cannot be completely filled into the custom block. Where the fill requirement refers to the fill word count requirement.

It will be appreciated that if the custom content data is not fully populated into the custom tile, it may result in the custom content data being scattered outside the custom tile, not only affecting the user's sense of viewing the file data, but also overlapping other data, resulting in file confusion, and in this case custom calibration information may be generated to calibrate the corresponding custom content data.

S42, responding to the customized calibration information, and acquiring a customized format corresponding to the customized demand information of the customized content data.

During calibration, the text format of the customized content data or the customized block can be correspondingly adjusted according to the customized format of the customized content data, so that the customized block can meet the filling requirement of the customized content data.

And S43, if the customized format contains font customized information, performing block adjustment on the customized block to obtain an adjustment block, and filling the customized content data into the adjustment block to obtain first calibration data.

It will be appreciated that if font customization information is present in the customization format, indicating that the user is required to customize the font format of the content data, in this case, the custom tile may be adjusted such that the adjusted tile may meet the filling requirements of the custom content data.

The specific implementation manner of step S43 based on the above embodiment may be:

S431, if the customized format has font customization information, acquiring the number of content words which are not filled into the customized block in the customized content data.

For example, if the number of filler words of the custom block is 300 words and the custom content data has 400 words, the number of content words is 100 words.

S432, obtaining a distance coefficient according to the ratio of the content word number to the reference word number, and obtaining a delay distance based on the distance coefficient and the reference distance.

It can be understood that the more the number of words in the content is, the more the number of words in the custom content data that cannot be filled is, so that when the custom block is adjusted, the area of the custom block can be increased, and the area-increased adjustment block can meet the filling requirement of the custom content data.

When the customized block is enlarged, the customized block can be processed by the extending distance, and the corresponding frame wire of the customized block is extended by the extending distance, so that the effect of enlarging the customized block is achieved.

In practical applications, the reference word number and the reference distance may be set correspondingly, for example, when the reference word number is x, the reference distance may be an extended reference distance y set correspondingly to the word number.

S433, carrying out extension processing on the frame wires positioned in the preset arrangement direction in the customized block according to the preset arrangement direction and the extension distance to obtain extension frame wires, and updating the customized rectangular frame corresponding to the customized block according to the extension frame wires to obtain an adjusted rectangular frame.

Specifically, when the preset arrangement direction is from top to bottom, the frame wires on the left and right sides of the customized rectangular frame corresponding to the customized block can be extended downwards according to the extending distance, then a new adjusted rectangular frame is generated according to the extended frame wires, and the original lower frame wire for sealing the left and right frame wires is deleted.

In some embodiments, the preset pixel value may also be called to update the adjusted rectangular frame, so that the user can learn that the rectangular frame is adjusted.

In addition, after the adjustment rectangular frame is obtained, the custom template can be updated and adjusted through the following steps:

and acquiring the customized rectangular frame positioned below the adjustment rectangular frame in the preset arrangement direction as a movable rectangular frame, moving the movable rectangular frame according to the preset arrangement direction and the extension distance, and determining the area corresponding to the moved movable rectangular frame as a customized block after movement update.

It can be understood that, because other customized rectangular frames may be arranged below the adjusted rectangular frames, in order to arrange the customized rectangular frames regularly, the customized rectangular frames below the adjusted rectangular frames may be correspondingly moved and adjusted according to the preset arrangement direction and the extension distance.

S434, determining the area corresponding to the adjustment rectangular frame as an adjustment block, and filling the customized content data into the adjustment block to obtain first calibration data.

After the rectangular frame is adjusted, the corresponding area can be used as an adjusting block, and the customized content data is filled into the adjusting block, so that the adjusting block can meet the filling requirement of the customized content data.

And S44, if the customized format does not have the font customization information, performing font adjustment on the customized content data to obtain calibration content data, and filling the calibration content data into the customized block to obtain second calibration data.

It will be appreciated that if no font customization information is present in the customization format, indicating that the user is not required to customize the font format of the content data, in this case, the font of the content data may be reduced, so that the adjusted calibration content data may be completely filled into the custom tile.

The specific implementation manner of step S44 based on the above embodiment may be:

S441, if the customized format does not have the font customization information, a preset reduced list is called, wherein the preset reduced list comprises a plurality of preset reduced scales.

In practical application, the preset reduction ratios in the preset reduction list may be preset by a worker, and each preset reduction ratio may have a corresponding word size. In the arrangement, the respective preset reduction ratios may be arranged from large to small in proportion.

S442, sequentially obtaining the corresponding preset reduction proportion from large to small as a target reduction proportion, and reducing and adjusting the fonts of the customized content data to obtain reduced content data.

It can be understood that the font of the customized content data is reduced from large to small, so that the customized block can be displayed with the largest word size as much as possible while meeting the filling requirement of the reduced content data, and the user can conveniently view the writing data.

S443, stopping font reduction processing of the reduced content data when the customized block meets the filling requirement of the reduced content data, and filling the reduced content data into the customized block to obtain second calibration data.

When the custom block can be completely filled with the reduced content data, the further reduction of the reduced content data can be stopped and the reduced content data can be filled into the custom block.

In this way, the custom block can be completely filled by adjusting the font of the custom content data.

S45, performing custom data calibration on the corresponding custom content data according to the first calibration data or the second calibration data to obtain calibration file data, and sending the calibration file data to the demand end.

In some embodiments, the customized rectangular box may be deleted to obtain full text file data, and the file data and the calibration file data may be sent to the user together, so that the user may check the content in the customized block.

The calibration file data obtained through the method is regular in arrangement, and a user can conveniently check writing data.

Referring to fig. 3, a schematic structural diagram of an AI writing system based on a scalable model according to an embodiment of the present invention includes:

the configuration module is used for receiving a customization request of a demand end, acquiring block configuration information of the demand end based on the customization request, and updating an initial template to obtain an expandable template;

the demand module is used for acquiring the customization demand information of the demand end for the corresponding customization block in the expandable template, wherein the customization demand information comprises necessary customization information and unnecessary customization information, and the expandable template is updated according to the customization demand information to obtain the customization template;

the customization module is used for receiving a writing request of the demand end, responding to the writing request, acquiring customized content data of a corresponding customized block based on customization demand information of the customization template, and updating the customization template according to the customized content data to obtain customized file data;

and the calibration module is used for retrieving an automatic calibration model to perform custom data calibration on custom content data in the custom file data to obtain calibration file data, and transmitting the calibration file data to the demand end.

The apparatus of the embodiment shown in fig. 3 may be correspondingly used to perform the steps in the embodiment of the method shown in fig. 1, and the implementation principle and technical effects are similar, and are not repeated here.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.

Claims (6)

1. An AI writing method based on a scalable model is characterized by comprising the following steps:

Receiving a customization request of a demand end, acquiring block configuration information of the demand end based on the customization request, and updating an initial template to obtain an expandable template;

Obtaining the customization demand information of the demand end for the corresponding customization block in the expandable template, wherein the customization demand information comprises necessary customization information and unnecessary customization information, the expandable template is updated according to the customization demand information to obtain the customization template, the necessary customization information at least comprises a customization writing direction, and the unnecessary customization information comprises demand information selectively input by a user based on unnecessary demand bars;

receiving a writing request of the demand end, responding to the writing request, acquiring customized content data of a corresponding customized block based on customized demand information of the customized template, and updating the customized template according to the customized content data to obtain customized file data;

retrieving an automatic calibration model to calibrate the customized content data in the customized file data to obtain calibration file data, and sending the calibration file data to the demand end;

Receiving a writing request of the demand end, responding to the writing request, acquiring customized content data of a corresponding customized block based on customized demand information of the customized template, updating the customized template according to the customized content data to obtain customized file data, and comprising the following steps:

Receiving a writing request of the demand end, and arranging each customized block in the customized template according to a preset arrangement direction to obtain a customized block sequence;

According to the arrangement sequence of each customized block in the customized block sequence, arranging the customized demand information corresponding to each customized block to obtain a customized demand sequence;

Generating demand content data according to each piece of customization demand information, and arranging the demand content data corresponding to each piece of customization demand information based on the customization demand sequence to obtain a demand content sequence;

Sequentially filling each piece of required content data in the required content sequence into corresponding customized blocks in the customized block sequence to obtain customized content data corresponding to the corresponding customized blocks, and updating the format of the customized content data according to a customized format corresponding to the corresponding customized required information;

updating the customized template according to the customized content data to obtain customized file data;

Retrieving an automatic calibration model to calibrate custom content data in the custom file data to obtain calibration file data, and sending the calibration file data to the demand end, wherein the method comprises the following steps:

when the automatic calibration model is called and the custom block fails to meet the filling requirement of the custom content data, custom calibration information is generated;

responding to the customized calibration information, and acquiring a customized format corresponding to customized demand information of the corresponding customized content data;

If the customized format contains font customization information, performing block adjustment on the customized block to obtain an adjustment block, and filling the customized content data into the adjustment block to obtain first calibration data;

If the customized format does not have the font customization information, performing font adjustment on the customized content data to obtain calibration content data, and filling the calibration content data into the customized block to obtain second calibration data;

Performing custom data calibration on the corresponding custom content data according to the first calibration data or the second calibration data to obtain calibration file data, and sending the calibration file data to the demand end;

If the customized format has the font customization information, performing block adjustment on the customized block to obtain an adjustment block, and filling the customized content data into the adjustment block to obtain first calibration data, wherein the method comprises the following steps:

If the customized format contains font customization information, acquiring the number of content words which cannot be filled into the customized block in the customized content data;

obtaining a distance coefficient according to the ratio of the content word number to the reference word number, and obtaining an extension distance based on the distance coefficient and the reference distance;

performing extension processing on the frame wires positioned in the preset arrangement direction in the customized block according to the preset arrangement direction and the extension distance to obtain extension frame wires, and updating the customized rectangular frame corresponding to the customized block according to the extension frame wires to obtain an adjusted rectangular frame;

Determining an area corresponding to the adjustment rectangular frame as an adjustment block, and filling the customized content data into the adjustment block to obtain first calibration data;

If the customized format does not have the font customization information, performing font adjustment on the customized content data to obtain calibration content data, and filling the calibration content data into the customized block to obtain second calibration data, wherein the method comprises the following steps:

If the customized format does not contain the font customized information, a preset reduced list is called, wherein the preset reduced list comprises a plurality of preset reduced scales;

Sequentially acquiring corresponding preset reduction ratios from large to small as target reduction ratios, and reducing and adjusting fonts of the customized content data to obtain reduced content data;

Stopping font reduction processing of the reduced content data when the customized block meets the filling requirement of the reduced content data, and filling the reduced content data into the customized block to obtain second calibration data;

after the adjustment rectangular frame is obtained, the method further comprises the following steps of:

and acquiring the customized rectangular frame positioned below the adjustment rectangular frame in the preset arrangement direction as a movable rectangular frame, moving the movable rectangular frame according to the preset arrangement direction and the extension distance, and determining the area corresponding to the moved movable rectangular frame as a customized block after movement update.

2. The method of claim 1, wherein the step of determining the position of the substrate comprises,

Receiving a customization request of a demand end, acquiring block configuration information of the demand end based on the customization request, and updating an initial template to obtain an expandable template, wherein the method comprises the following steps:

receiving a customization request of the demand end, calling the initial template and sending the initial template to the demand end;

acquiring specification information input by the demand end to a preset rectangular frame, and updating the specification of the preset rectangular frame according to the specification information to obtain a customized rectangular frame;

Receiving positioning information of the demand end on the customized rectangular frame in the initial template, filling the customized rectangular frame to a corresponding position in the initial template according to the positioning information, and taking a region corresponding to the customized rectangular frame as a customized block;

And obtaining block configuration information according to the customized blocks, and updating the initial template based on the block configuration information to obtain the expandable template.

3. The method of claim 1, wherein the step of determining the position of the substrate comprises,

Obtaining the customization demand information of the demand end for the corresponding customization block in the expandable template, wherein the customization demand information comprises necessary customization information and unnecessary customization information, and updating the expandable template according to the customization demand information to obtain the customization template, and the customization demand information comprises the following steps:

Receiving a demand customization request of the demand end for a corresponding customization block in the expandable template, calling a necessary demand column and a non-necessary demand column corresponding to the corresponding customization block, and sending the necessary demand column and the non-necessary demand column to the demand end;

acquiring necessary customization information input by the demand end based on the necessary demand column and unnecessary customization information input by the unnecessary demand column, and acquiring customization demand information according to the necessary customization information and the unnecessary customization information;

and updating the expandable template according to the customization demand information to obtain a customization template.

4. The method of claim 3, wherein the step of,

The obtaining the necessary customization information input by the demand end based on the necessary demand column and the unnecessary customization information input by the unnecessary demand column comprises the following steps:

Acquiring the requirement information of the requirement end for inputting all necessary setting data in the necessary requirement column to obtain necessary customization information;

and determining unnecessary setting data selected by the demand end in the unnecessary demand column as target setting data, and acquiring demand information input by the demand end on the target setting data to obtain unnecessary customized information.

5. The method of claim 1, wherein the step of determining the position of the substrate comprises,

Receiving a writing request of the demand end, and arranging each customized block in the customized template according to a preset arrangement direction to obtain a customized block sequence, wherein the method comprises the following steps:

Acquiring block center points corresponding to the customized blocks in the customized template;

And arranging the customized blocks according to the arrangement positions of the block center points in the preset arrangement direction to obtain a customized block sequence.

6. An AI authoring system based on a scalable model, comprising:

the configuration module is used for receiving a customization request of a demand end, acquiring block configuration information of the demand end based on the customization request, and updating an initial template to obtain an expandable template;

The demand module is used for acquiring the customization demand information of the demand end for the corresponding customization block in the expandable template, wherein the customization demand information comprises necessary customization information and unnecessary customization information, the expandable template is updated according to the customization demand information to obtain the customization template, the necessary customization information at least comprises a customization writing direction, and the unnecessary customization information comprises demand information selectively input by a user based on an unnecessary demand column;

the customization module is used for receiving a writing request of the demand end, responding to the writing request, acquiring customized content data of a corresponding customized block based on customization demand information of the customization template, and updating the customization template according to the customized content data to obtain customized file data;

The calibration module is used for retrieving an automatic calibration model to perform custom data calibration on custom content data in the custom file data to obtain calibration file data, and sending the calibration file data to the demand end;

Receiving a writing request of the demand end, responding to the writing request, acquiring customized content data of a corresponding customized block based on customized demand information of the customized template, updating the customized template according to the customized content data to obtain customized file data, and comprising the following steps:

Receiving a writing request of the demand end, and arranging each customized block in the customized template according to a preset arrangement direction to obtain a customized block sequence;

According to the arrangement sequence of each customized block in the customized block sequence, arranging the customized demand information corresponding to each customized block to obtain a customized demand sequence;

Generating demand content data according to each piece of customization demand information, and arranging the demand content data corresponding to each piece of customization demand information based on the customization demand sequence to obtain a demand content sequence;

Sequentially filling each piece of required content data in the required content sequence into corresponding customized blocks in the customized block sequence to obtain customized content data corresponding to the corresponding customized blocks, and updating the format of the customized content data according to a customized format corresponding to the corresponding customized required information;

updating the customized template according to the customized content data to obtain customized file data;

Retrieving an automatic calibration model to calibrate custom content data in the custom file data to obtain calibration file data, and sending the calibration file data to the demand end, wherein the method comprises the following steps:

when the automatic calibration model is called and the custom block fails to meet the filling requirement of the custom content data, custom calibration information is generated;

responding to the customized calibration information, and acquiring a customized format corresponding to customized demand information of the corresponding customized content data;

If the customized format contains font customization information, performing block adjustment on the customized block to obtain an adjustment block, and filling the customized content data into the adjustment block to obtain first calibration data;

If the customized format does not have the font customization information, performing font adjustment on the customized content data to obtain calibration content data, and filling the calibration content data into the customized block to obtain second calibration data;

Performing custom data calibration on the corresponding custom content data according to the first calibration data or the second calibration data to obtain calibration file data, and sending the calibration file data to the demand end;

If the customized format has the font customization information, performing block adjustment on the customized block to obtain an adjustment block, and filling the customized content data into the adjustment block to obtain first calibration data, wherein the method comprises the following steps:

If the customized format contains font customization information, acquiring the number of content words which cannot be filled into the customized block in the customized content data;

obtaining a distance coefficient according to the ratio of the content word number to the reference word number, and obtaining an extension distance based on the distance coefficient and the reference distance;

performing extension processing on the frame wires positioned in the preset arrangement direction in the customized block according to the preset arrangement direction and the extension distance to obtain extension frame wires, and updating the customized rectangular frame corresponding to the customized block according to the extension frame wires to obtain an adjusted rectangular frame;

Determining an area corresponding to the adjustment rectangular frame as an adjustment block, and filling the customized content data into the adjustment block to obtain first calibration data;

If the customized format does not have the font customization information, performing font adjustment on the customized content data to obtain calibration content data, and filling the calibration content data into the customized block to obtain second calibration data, wherein the method comprises the following steps:

If the customized format does not contain the font customized information, a preset reduced list is called, wherein the preset reduced list comprises a plurality of preset reduced scales;

Sequentially acquiring corresponding preset reduction ratios from large to small as target reduction ratios, and reducing and adjusting fonts of the customized content data to obtain reduced content data;

Stopping font reduction processing of the reduced content data when the customized block meets the filling requirement of the reduced content data, and filling the reduced content data into the customized block to obtain second calibration data;

after the adjustment rectangular frame is obtained, the method further comprises the following steps of:

and acquiring the customized rectangular frame positioned below the adjustment rectangular frame in the preset arrangement direction as a movable rectangular frame, moving the movable rectangular frame according to the preset arrangement direction and the extension distance, and determining the area corresponding to the moved movable rectangular frame as a customized block after movement update.