The automated generation of AI comics, which relies on special aids, should start with the generation of audio after getting a sales license in the normal order of operations, and does not require the manual creation of novel subplot scripts and corresponding screen cues.
This tutorial is only for learning the principles involved. If you want to practice it quickly, just find the tool yourself.
Preparation of novel material
Follow the previous tutorial Demystifying AI Comic Book Narration Video from Principle to Practice (Acquiring and Rewriting Fiction) Study and you'll get three copies of the material:
reputation as calligrapher
The Fairy's Rebellion: A Self-Help Manual for Abusive Women.
beginning of a novel
Fairy White Lotus, white and beautiful with long legs, once crossed over into a sadistic female match, her road of reversal has just begun! White Lotus, the daughter of the Heavenly Emperor, was originally the favorite of the immortal world, but was accidentally involved in a sadistic plot. Her fiancé, Ling Feng, is a genius in the immortal world, but he is dissatisfied with her marriage. Bai Lian decides to rewrite her destiny, starting with a jar of peach blossom brew, her fun daily routine with the young fairy E Li Li, to the strange encounters in the secret realm, stepping out in her own style every step of the way. When the legend of the dragon jade pendant appears in the world, how will Bai Lian's encounter with the mysterious little demon, Qing Yi, turn her life upside down? In this abusive story, how will Immortal White Lotus move step by step towards her own splendor?
adaptation of novel
I'm dressed as a fairy.
Not the fairy of mortal legend, but the real kind that flies around in the sky. The Emperor of Heaven is my father and the Queen of Heaven is my mother, so I was born a fairy, with white skin and long legs, and full of fairy spirit.
The days of a fairy are simply too comfortable. Getting high, reading books, and playing with the little fairies in their own fairy palace all day long is awesome.
Unfortunately, the good times didn't last long before something happened.
That afternoon, my palace's little fairy servant, Li Li, came running in with great haste, "White Lotus Supreme Immortal! White Lotus!"
"Hear hear hear, stop yelling." I'm still not quite used to the name, even though it's been months since I crossed over.
"Immortal Lingfeng has returned from his descent into the mortal world to cross the tribulation, let's go and visit."
"Who is Immortal Lingfeng?"
"Supreme Immortal are you sleepy, that's your fiancé!"
"Huh??"
It was then that I remembered that I seem to have a fiancé. Why does the name Ling Feng sound so familiar, it's exactly the same as the hero of the novel I read before I crossed over?
After some inquiries, I finally figured out the situation: I, White Lotus, am the female counterpart in this abusive text.
The main character, Ling Feng, is an upper immortal in the immortal world, reaching the position of upper immortal in 20,000 years, a rare talent in the immortal world. Although I am also an Upper Immortal, unlike him who relied on his own skills to cultivate, I am an Immortal Second Generation, born as an Upper Immortal.
My Heavenly Emperor dad saw Ling Feng's talent, and at my urging he hit it off with Ling Feng's master, Ancestor Bodhi, and set up a marriage between the two of us.
But Lingfeng was not at all satisfied with the marriage, and also had a big fight with the Bodhi ancestor, and finally parted on bad terms, turning around and going down to the mortal world to experience the calamity.
I counted the time, this time, the robbery, Ling Feng reincarnated into a cultivation sect, and the female lead is his personal disciple. He didn't go back to the immortal world immediately after the calamity, but stayed on earth for a few more decades and waited until the female lead, Qian Yue, succeeded in ascending before they came back together.
The original owner originally had a good feeling for Ling Feng as a disciple as well, but when she found out that Ling Feng's feelings for Qian Yue were more than just master and disciple, she started to go crazy with her death wish, trying to find ways to break up the two of them.
In the end she couldn't help but be jealous and pushed Qian Yue off the Immortal Exorcism Platform, almost sending people's souls flying. After the truth came out she was expelled from the Immortal Realm and was also killed by Ling Feng himself and her soul was dispersed.
The thought of it gave me a terrible headache. I don't want to meet someone who's going to beat the hell out of me later.
But in order not to collapse my persona, I was still dragged by Lili to the Lingfeng Palace for the banquet.
Lingfeng Palace was built next to Peach Blossom Valley, and it was a good time to enjoy the flowers. I swiped my face to enter the palace gate and slipped out of the banquet in the twinkling of an eye. Passing by the liquor place, I smoothed down an altar of Peach Blossom Brew and ran to the Peach Blossom Valley to drink and enjoy the flowers.
Peach Blossom Brewing really lives up to its name, it's very good to drink, but it has a strong aftertaste. In a while I got sleepy, and found a tree to lie down and squint for a while.
In a daze, I heard a movement, I thought it was Ri Ri looking for me, so I poked my head out of the tree and prepared to pamper myself. But it wasn't Ri Ri who came, it was the heroine Chiyuki.
Qian Yue obviously didn't expect this someone and froze, looking at me with tearstains still on her face. It was then that I remembered the plot written in the novel where the heroine was ridiculed for her looks when she first came to the immortal world and ran out in tears.
Lol, it's childish to make fun of looks when you're thousands and thousands of years old immortal.
As a qualified White Lotus, ah no, White Lotus Upper Immortal, how could I not care when I saw a cute girl being bullied and crying? I snapped my fingers and summoned the flower demons, making them dance around Chiyuki. Qian Yue had never seen this battle before, her face instantly turned red.
She looked at me cautiously, and I smiled at her, hearing RiRi calling out to me in the distance, and beckoned a cloud to go home with RiRi.
Within a few days, my Heavenly Stepmother came to visit me.
"Son, have you seen Ling Feng?" The Heavenly Empress smiled extraordinarily kindly.
"Ah... No." I slipped out early and didn't see anyone at all.
"This time he came back from the calamity, his cultivation has made great progress again, your father said that he has a great future, you and him..."
Gotta be, it's a rush to get married.
process script
Start processing the script with the content of the adapted novel. Processing the script of the novel is carried out in three steps: organizing the original text of the novel into segments - converting the screen keywords in segments - supplementing the description of characters (roles) in the screen keywords.
Processing script process will be in accordance with the logic of an AI batch drawing tool, here does not recommend any tool, only for the demonstration of teaching.
The final goal of the processing script is to get the following two tables: (because these two tables are imported into the import file of a tool for batch charting...)
1. Split the original text of the novel into "descriptions of the original text".
Regarding the problem of splitting the description of the original text and the corresponding screen keywords that cannot be processed together: the reason is that the input and output texts are too long, and the large model is unstable, so it is split into two steps to be carried out separately. Of course, the use of "workflow" or "function call" can be a one-step solution to this problem, but it is not suitable for ordinary people.
input
ChatGPT, KIMI, Claude, are recommended.
The original novel is below: === {Paste the original text of the novel} === You are good at segmenting the original novel text based on professional movie footage to use as a movie script. Ensure that each segment of the original text conforms to the movie script shot sense. Merge or split segments of the original novel content for the same shots. Output a table with two columns: Number | Original Text Segmentation *Note: The entire original novel text in the three equal signs '====' must be segmented, with no breaks in the task or omission of the original novel text allowed for any reason.
exports
The two tables I mentioned above can now be filled in with the first column first:
Large models do the segmentation thing, it is always very unstable, especially the single column output, if you add a column to generate the screen keywords will be much better, but the output is too long more unstable.
So here's another approach that's more stable and can be added to automated workflows: theBuckle tutorial: using the "code node" will be a long text for segmentation
2. Generate screen keywords
input
Recommendations ChatGPT, Claude
Original segmented table: === {Two-column table in makrdown format output from previous step} === # According to the deduce a complete picture, imagine the missing or implied information, then carry out the sub-scene description, the sub-scene description is composed of a group of comma "," split by the English TAG, known as the "TAG group". ## TAG group requirements: 1. Do not use pronouns for each sub-TAG. Avoid these words: "in a, the, with, of, the, an, and, is, by, of." 2. 2. Each set of TAGs contains the following categories: 2.1 Expression: facial expressions 2.2 Action: a detailed description of a character's movements in a scene. 2.3 Scene: a small scene within the scope of the location on which the camera is focused, the small scene is a dynamic or static description of the character/scenery, not limited to: a guest room, in front of a mirror, on a bench, on a bed, in a garden, a handsome young man standing by his side, a group of young girls, a maid of honor, a fish swimming in the water, a red sun, a white cloud in the sky (more than one sub-TAG is allowed to be generated) 2.4 Location: describes the location, the scene, not limited to: palace, airport, coffee shop, airport, road, etc. 2.5 Lens: lens distance and perspective, according to the location, scene, action, give the near and far view and lens perspective, TAG group should be reasonably interspersed with different lens distance // 2.3 Scenes target potential information in the sub-scene descriptions, allowing multiple TAGs to be generated // Each sub-TAG appears in the scene, the location of the word style should be consistent with other shots 3. Do not display the category name in the TAG group output. 4. Generate a "TAG group" corresponding to the original segment by number. ## List of Main Character Names - White Lotus - Li Li - None // Only one name can be selected from the in the three-column table. // If you reason that a TAG group is not suitable for a main character's name to appear in a subplot description, write "None" in the "Main Character Name" field of the three-column table. ## Outputs a table with three columns: number|TAG group|protagonist name
This program generates a cartoon for each sentence, which actually makes little sense, especially when using MJ, which is more costly, and can be used to combine multiple narrations into one and generate a single drawing, after all, people are mainly listening, not looking at drawings. Please be adjusted for your creative focus.
The previously written cue word seemed very clearly constructed and not necessary, so this time a new cue word was written in the order of the natural description, and the effect was identical.
Recommended models: Claude, deepseek (code assistant version), ChatGPT, Smart Spectrum
1. Because the logic is too complex, it is recommended to add your own comic style "example" screen prompt words, otherwise not every generation is absolutely stable.
2. Regarding the main character name recognition, it needs to be tested and modified on its own. According to the current method to achieve character consistency (I want to fix the character based on the cue word), it is suggested to force to add a unique character to the third column of the table.
3. Retaining the number is to force the model not to produce forgetfulness, to complete the task at once, and to facilitate troubleshooting.
Alternative: one-step generation of a novel narration plus a corresponding screen scheme:
- This program presupposes that the original novels are not too long to be processed at a time (three or four thousand words at a time is not too much of a problem when the right model is chosen)
- The advantage is that as long as you choose the character style how to generate a more consistent "mini-model", you can make the screen structure more rich.
You are a professional AI painting cue word expert, I will give you a sentence and you need to help me generate a cue word suitable for AI painting based on this sentence with the following rules: 1. Give you the content of the novel fragment and then the meaning of the same case of rich details, and make him into a sense of description of the expression of the picture. 2 need to determine the text said the background time, for example: ancient China, modern, modern, future, etc., if the paragraph does not have obvious tips, then you can judge according to the context of the paragraph 3. If it is in the first person point of view, then the text is transformed into a picture description; the rules are as follows: can not change the meaning of the sentence, can not be ignored, can not be fabricated, to be logical, retain the character's name, delete the character's dialog. The main body of the characters according to the semantic judgment using: a man, a woman, a boy, a girl, a little boy, a little girl, a middle-aged woman, a middle-aged man and other words to describe; the main body placed in the forefront of the front, the action description connected to the back, the background or the scene description placed in the middle of the overall modification to be placed on the last. 4. the paragraph contains characters, according to the times and background of the description of people corresponding to the characters wearing 5. If the description is just a scene, there is no need to characterize the main body description 6. According to the text of the passage, please provide the relative scene suggestive words. 7. According to the text of the subject of the human object or environmental description, with the movie camera thinking to determine the picture is suitable for the use of that kind of scene (for example: close-up, close-up, panoramic, medium, distant) and add to the prompt word 8. According to the environment of the text to add the atmosphere of the descriptors such as: cool/warm tones, sunny, well-lit 9. Generate prompt words that match the meaning of the sentence, not just make up nonsense. 10. The descriptors you generate can't exceed 60 words, please strictly observe The following are a few cases please refer to: Case input: The morning sunlight slanted brightly through the large window and fell on the desks, and outside the tall aspen trees Case Output: Classroom, Morning, Tree Projection, Medium Ground, Warm Tone, Modern Campus, Depth of Field, Shadow Spots, Halos Case Input: Xu Yang looked around, and if memory serves, it was now her first class meeting after entering the university. Case Output: A female college student in a casual top and jeans looking around in the classroom, close-up, depth of field, warmth, modern campus, natural light, halo Final output in three columns: Number|Original Text|Output Screen The following are the inputs to be output: ===== {paste the two original numbered and segmented columns, why they are numbered has already been stated} ===
PS: Forget the alternative, look at the output you get from the default input command...
exports
At this point you get a two-column table:
Look carefully at the above picture keywords in the lack of detailed character description, deliberately did not add, because I want to use later AI batch map software can be set up separately in the character description words. The following:
Directly let the big model in the picture descriptor to add the character (character) descriptor is very troublesome, but also can not unify the image, so the AI batch drawing software will be abstracted out of the character description of a separate management.
We have added a third column in the EXCEL table, indicating whether to activate the "Description of a character" in the screen, 0=not activated, 1=activated, activated means that when generating the picture, to put theCharacter descriptors inserted into picture descriptorsCenter.
Why did you add the C and D examples? Because the AI batch mapping software I use supports importing by table.... ...C means activate character 1, D means activate character 2, leave blank means don't activate character traits.
Quick formula for labeling characters:
Simulate how the AI Batch Out software controls the characters:
Example: "Empty" indicates that a characterization is not activated:
Example: "1" indicates activation of a character description:
3. Creating character description keywords
Where to write the character description that needs to be activated? Different AI batch image software operation process is different, I just add it to the EXCEL template here.
Start creating personalized keywords for the different "activated" characters, either by typing in the context of the novel and letting ChatGPT help you visualize it, or by tweaking it yourself in a drawing tool. I've made Bailian's hair a different color than her clothes, and I've made Ri Ri wear a ball head to create a difference.
4.Complete EXCEL file downloads
ultimate
We have done the novel script text processing work by hand, and also reproduced the underlying logic of most AI novel mapping tools. Although there are differences in the execution logic of each mapping tool, you can understand the principle from it, and here again, we do not recommend any tool.
The next article we will enter the formal image generation to explain, before we mainly understand the principle of which, it will be very difficult to learn, the subsequent teaching use to supporting tools, learning will be very simple.
Forewarning: in the next post we will explain how to generate videos ~ so much padding, finally come to the point... Generate video is the easiest of all the teaching, the focus is to find a smooth tool, there is a wealth of material on hand. So the next lesson we will give you to organize a number of commonly used resources to generate novels, remember to pay attention to the collection Oh.
Demystifying AI Comics Article Series Table of Contents
- Demystifying AI Comic Interpretation Video from Principle to Practice (Principles)
- Demystifying AI Comic Explainer Videos from Principle to Practice (Sub-Commissioning Channels Chapter)
- Demystifying AI Comic Book Narration Video from Principle to Practice (Acquiring and Rewriting Fiction)
- Demystifying AI Comic Book Narration Video from Principle to Practice (Making Novel Scripts)
- Demystifying AI Manga Narration Video from Principle to Practice (Composite Video)