[Music] Good morning children How are you daddy brought food for all of you let's pray before eating [Music] this is a video with 3D animation style What I made was inspired by a free nutritious meal video that some time ago was circulating on social media Not only inspired, I even intentionally made the story according to the one in the original video If you may have never watched the original video made by the Ministry of Communication and Information you can just search on YouTube free nutritious meal video you will definitely find it. but don't compare between my creation as an ordinary citizen and the creation of the Ministry of Communication and Information, it must be different even though the story is similar Because I made it by copying a video made by this Ministry. but here I recreate it the visuals are different my purpose in making this video just to practice how to make a 3D animation style video using Ai then I can share the tutorial to those of you who might want to make animated film content I hope this video can be useful because I made it using the Dzine Ai tool a multifunctional tool with many features Not only can you make images and videos This tool also has a custom character training feature so we can make consistent character of people in many scenes In addition, there are many other editing features that can be used, for example in my previous video on how to apply a hijab product photo to a model made by Ai and many more features in this Dzine Ai tool for the Dzine Ai link I've written in description column, you can just check there let's start this tutorial right away usually the first stage I always create a story idea first with the help of chatgpt now because of the video story idea I adjusted it to the original free lunch video then the first stage I immediately asked chat PT to make a scene prompt the way is like this. here I immediately uploaded a free nutritious food video file here I have uploaded then I added a Prompt like this I have attached an animated video file blah blah blah blah blah and I asked ChatGPT to make a Prompt the scene image directly like that. Well, below I will give you chatgpt information or details to be considered in making image prompts from the video This is an example of the core scene I will just give you the outline of the video showing a man comes to a school carrying a stack of lunch boxes then goes into the classroom and so on, the second one make sure Prompt produces composition details and the same feel as the reference video then I give information too to chatgpt that I will remake this video with Ai text to Image tools then image to video. pay attention to each image Prompt contains elements of photography which is cinematic as follows subject, scene, composition, lighting and others and so on then chatgpt answers like this. he starts analyzing the video file that we uploaded. here chatgpt said he would analyze the frames to determine the main visual elements such as lighting, color, composition, and others, until finally chatgpt can directly give us a scene prompt according to what we want. well this is the result here is the Prompt, the first, the man came to school bringing food then entered the class greeted by students with applause. the man distributed food to the students. praying together before eating children are happy blah blah blah men end with namaste then clap their hands then here I write the details that I want to pour in the image prompts you can read this yourself, okay? Well For example, for all Prompts produces cartoon animation style images then the Prompt that displays the face of the main character is a man carrying food then chagpt must consistently write the character traits Well here I write the characteristics are roughly like this age over 50 years blah blah blah blah black hair, navy blue short-sleeved shirt to details like where to eat must be silver or chrome and don't forget this is also important all students wear white shirts then chatgpt revises the prompts earlier perfecting according to the details I added like this. Well here I feel these six scenes are still incomplete because these six scenes are just the main scenes. if we only make these six scenes the video will feel choppy and the story it won't be coherent, right? so we need add additional scenes then I ask here chatgpt to make 10 Prompt new additional scenes, so that the video is more coherent and interesting to watch Now here I give an additional example like a scene showing the main character arriving in front of the classroom door another example here, there is a male main character walking among the students then there is a close up scene of the male character praying. so in this process, if we want add more details that we want we just go back into chatgpt Now then chatgpt rewrites all Prompts, totaling 6 + 10 additional Prompts, so there are 16 Prompts in total So if you feel that everything is complete This Prom is in accordance with our wishes, then actually this is finished, we can generate this Prompt into an image we enter the stage next namely creating a dataset in the form of a 3D cardboard image the main character in this video so before we start generating the scene image we first create a dataset in the form of a cartoon image of the main character which will often appear in this video for us to use as custom character training material so in this video I made this 3d cartoon character comes from original photos, so I collected all photos of the main characters that will be taken in this video, you can prepare where they come from whether it's my own photos or if you want to take them from the internet because if in Dzine written We are told to collect 1 to 40 image files to make a data set. Well here I around maybe not up to 20 I've collected these photos then we go straight to the image to Image feature in Dzine Ai. now the display is like this for the Ai design tools. Now in the image to Image feature, we set the settings are like this. we choose 3D and this is specific in Qute 3D, for style intensity we choose the weak 0.4 structure match choose the most strict then we choose activate Face match and high quality, that's all, then we upload the photo we want to convert into a cartoon image, after the upload is complete we can immediately click generate and the results are like this we try again for other images we upload here, then let's make sure the settings are all the same if it's ready, we just generate it the result is like this Now here we just choose which one is the best in our opinion Now to tidy up the results of this image, we delete the background part, so that it doesn't disturb the training results later because this background is too busy we just place it on Canvas, then we remove the background like this when it's finished, then we can directly download the results like this Here are some images that I have deleted the background. we continue directly to the custom character training process here we use the consistent character feature then we make a new character. Well here we can enter the name of our character Let's say here, the cartoon president then we upload all the images that we have prepared earlier we just wait for the upload process it's done, we can start right away train process. Well here it says the duration approximately around 19 minutes. let's just wait until it's done. Well, when it's finished, our character results will appear on this list. we can continue directly generate Prompt into a scene image now when we choose our character he will automatically appear here this is the Prompt that appears automatically describing the characteristics of the character Now here I do slight adjustment from the character traits, to match the character traits I want here there is light skin, short Combed Black Hair close lip smile, then the character also wears a short-sleeved navy blue shirt now details like that we can adjust according to the character we want then in the Prompt column below, this is used to enter our scene Prompt which we generate from chatGPT Now here the character has been represented with the sentence cartoon president we add here, He is, next we go to chatGPT, for example this Prompt we start from the word standing, down, now for the word the one before standing he describes the character traits, while in Dzine earlier we've already made a custom character so we don't need to include the character details from this GPT chat because it's already represented by the cartoon president's sentence earlier. here we just copy it, then we paste it here. Now here you can see He is standing blah blah blah blah until finished, then the settings we choose 16 to 9. That's it. then we directly generate and the result is like this we just repeat these steps to generate all the scene images then we go to the process to convert this image into a video here I use Runway Now for this first scene The prompt is roughly like this tracking shot, the man walks forward, the camera follows from behind. then we generate and the result is like this Now let's try for another scene like this. here I use Prompt: Static shot the man is greeting, natural movement. we generate and the result like this. and this is for the video generate result for other scenes here I have included the prompt friends, believe it or not most of the Prompts I use for convert this image in Runway, that's almost all of it using the natural Movement Prompt with this Prompt alone I can when I use a more detailed Prompt sometimes the results are not better There are only a few scenes that need a More Detailed Prompt For example, He is Walking natural Movement we enter the stage of making a lipsync video using Runway act one to get lipsync results with character voices as we want which in this video I want to use a voice similar to Mr. Prabowo. the way I do it is by taking a voice sample from a YouTube video that shows him giving a speech or a greeting. then I process the audio sample file to the voice cloning tool here I use a free tool it's called vocloner. the method is like this here we open vocloner, then we enter the name of the voice character now in this section, we will upload the audio sample here I will play a little for the audio sample that I took from YouTube. like this. "with this step which will come into effect in March" well something like that. that's the original voice after we upload we immediately process submit Now when the cloning process is complete then the name of the voice character will appear here then we continue to the text to speech process changing the text into voice according to the character we have created now in this section, we enter the text that we want to change into character voice here then the format is MP3. then we generate speech. the result is like this "Good morning children, How are you, father brought food for all of you let's pray before eating" Well, I think this is pretty good, if for the color of the voice is very similar although indeed the intonation is a bit stiff then the next stage we prepare Driving video for us to use in Runway act one feature. so here I've prepared it, I'll record it close up of the face acting to say the words we generated earlier the result is like this. footage recording that I took. "good morning kids how are you" so this video I act saying the words we made earlier we try to be as similar as possible and then this video footage we put it into capcut at the same time as the character's voice that we made earlier we make the lip movements according to the voice that we made earlier so it's something like this after we arrange it, just adjust it the result is like this. "good morning kids how are you" and this is Driving video for the second scene "let's pray before eating" Now after we adjust the lip movements with the voice of the character we have created, the result is like this. "let's pray before eating" if it's okay, we just export this video then we save it then we bring the finished video to the act one feature on Runway at the top here, where we upload Driving video that we have made earlier "let's pray before eating" then in the bottom section here we upload for the character video that we have made we wait for the upload to finish then we immediately generate and the result is like this. "let's pray before eating" and for the second lipsync scene the result is like this. "good morning kids, how are you" before we go into video editing in capcut Actually there is a stage of determining the background music but here I have often taught my friends where to get free music easily or my friends can make it themselves through song making platform using Ai So I think friends there is no difficulty in getting music this can be taken from anywhere according to needs. Okay so I continue go straight to the video editing stage in capcut and here is the capcut PC display, guys you can see here, I will explain from the beginning in the first scene, here is text friends can design their own text there are many choices of effects and fonts in capcut plus animation to make it more interesting here there are footsteps in all parts When walking. there is the sound of footsteps well when starting to enter into the class, there is an additional sound effect like the sound of a crowd in a room like that. Well in this scene there are also additional sounds like children's voices, yes we have often heard this sound in the scene that shows food we can give it here, there is an effect friends can use it in this effect section there are many choices of what kind of effect you want plus there are sound effects Sparkling sound. continue. because this is still in class so still use sound effects like crowd noise in this section, there is also text but in Bubble form friends can use this in the Bubble menu here. continue. until the last scene there is also writing that we can set according to what we want it's like that again for how to edit videos depends on our taste and our respective creativity That's all for today's video tutorial I've explained from start to finish how to make 3D animation videos using the Ai tool I hope this video can help you who want to create animated video content because with the help of tools like Dzine Ai, making videos with 3D animation style is easier want to make various stories no matter how creative, all can be made with the help of Ai and what's even more special especially with the presence of tools like Dzine Ai now make more consistent characters in every scene becomes easier if for content purposes on social media this tool seems to be more than enough because the editing features are also very complete press the button like if you like with this video and if you have any questions, don't hesitate to comment If you want to ask more intensely you can join our discussion group on telegram the link has been written in the description column see you in [Music] next video