No timeline editing, no key frames, no talking. All done with AI. Grab the full training below. Unlimited videos, zero credits. Let me show you how easy it is to build this in video.ai.
And the thing you ordered like this is the video I’ll be showing you. One second while I’ll play this video. So, we’re going to show you today exactly how to build a video like this with Video Express. And the other cool part is that we’re going to bring it bring the whole thing together as well. So, let’s play this video real quick. And it has my voice. I’ll show you how you can add your voice into the video as well. And uh basically how to add these whole animation setup. Okay, step by step the whole thing. So, let’s play this first. You tap by now. The thing you ordered seems like it could be sitting in a warehouse thousands of miles away. and 2 days later it’s at your door. How you picture a worker sprinting down endless aisles to find your thing. But Amazon isn’t waiting for your order then scrambling to find it. It guessed long before you click its systems forecast what people in your region were likely to buy and moved inventory closer in advance. So when you tap by your package may already be in the same part of the country. It may be sitting in one of hundreds of fulfillment centers around the world. Inside, robots move shelves and products toward employees, cutting much of the walking out of the search. Amazon has deployed more than 1 million robots across its operations network, the world’s largest fleet of industrial mobile robots. A worker picks the right item and the order gets packed and labeled. Then it moves through a sortation center where packages are grouped by destination for the next leg. The package can’t linger. The network chooses its next route ground when it’s close. air when distance demands it. Some shipments travel through Amazon Air Cargo’s network of more than 100 aircraft operated by carrier partners, moving packages while you sleep. At a local delivery station, routing technology builds routes around package volume, address complexity, and neighborhood conditions. Then, delivery partners move from stop to stop, carrying hundreds of packages through the surrounding community. Zoom out and it stops looking like a store. Across 19 countries, Amazon’s delivery service partners deliver more than 20 million Amazon packages a day. It’s one machine coordinated across fulfillment, sortation, air, road, and doorstep delivery. So, fast shipping was never only about moving one box faster. The real trick is forecasting demand and placing inventory close enough that much of the journey has already happened before you click. All right. Pretty cool, right? So, this is what we like to call, if you will, walk style or let’s say an an explainer style video is a better way to say it. And the way, if you see this video was done, there’s two ways to do it. I’ll show you both ways. And I have a workflow for you that I built. You can go into video.ai/workflow to get it. And basically, this is going to open up a page that looks something like this. And when you go to that page and then here you’ll see right here this highlighted section on the top it’s going to have this you see this motion graphics animation workflow uh that move with a purpose. So this is the workflow and um so when you come here right now it’s for CHPd because what I’m doing is I’m going to be showing you some things you can do uh if you have a CHP plus account as well that is pretty cool uh with this process if uh per time permitting. We’re going to get to that as well. But I’m going to show you how this workflow uh basically does the whole process when you go there. So it’s going to open up this GPT that I built for you. And this GPT can also be used by free chat GPT accounts if you will. And then right underneath it, you’re going to see something like this, this little uh option. You can start with this. This is what we call a conversation starter. So when you click this um it’s going to start uh you see where it says turn this idea into a 60-second explainer video. Paste idea. So effectively there’s nothing happening here, right? So the better way to do it is uh basically we copy this this little thing. Okay? And then you paste it here. Okay? And then effectively you’re going to say for example how Amazon delivers packages in let’s say one day. So you type that in and then it’s basically going to do the thing right it’s going to generate it and you don’t need a pro account or you don’t need anything else. It can pretty much do this cuz I’ve done all the thinking for you. I’ve done all the process for you behind the scenes what is happening. Let me explain as well while this thing is doing its thing. Uh is effectively it learns. So let me write this down so you can understand right here on the right. So it’s learning exactly how the prompting structure works in video express. Okay. So uh it I basically taught it everything about video express this prompting style and everything the version three uh and then it also what it does is it also learns the prompting structure uh for in this particular case what I’ll show you how to do is a text to video option. The reason why I’ll show you text to video specifically is there’s a cool things you can do with text to video for vertical as well as 16×9. I’ll show you how to do the same process because I showed you this one. I show you how to do even um 9×6 as well the vertical format and then also what I’m going to be showing you what I did behind the scenes for this workflow if you go to that page. But the other thing that I did is give you the whole uh sequences, right? So it’s going to do scene one, right? and uh scene number one. It’s going to do scene number two. Okay, and on and on. Okay, it’s going to give you the scenes for this. And it’s also going to have the speech embedded into it. Okay, which is pretty cool. Uh so the idea is that in the prompt itself, we’re going to put speech because uh in video express as you know, you can actually make uh narration videos on the fly as well. And this could be your voice too. This is going to be a little bit extra steps if it’s going to be your voice, but it can be even an AI generated voice which is much faster. Uh so which I’m going to show you as well. So it’s doing all this stuff here. You see when it’s saying this stuff, this is what it’s doing behind the scenes. Okay. So uh obviously you can use a free account to be able to do this in Chai GPD or a pro account plus account whatever. It doesn’t really matter. So it’s doing all these for you to then basically copy and paste this these scenes into uh your video express account. and I’ll show you where to copy and paste it. Now, the other thing I’m going to do in the next step is if you have u the non-free account, let’s say the plus account, right, which is I think $20 a month or a go account if you’re in some parts of the world which is I think $8 a month or something. Let’s say you have one of these accounts. What they introduced a new feature which I’ll show you after this. It’s called the browser control. Okay. Um which is available uh in u the plus accounts right now. They’ll roll it out to free accounts later on. I don’t know. Uh but what this will do is the same structure here that we’ve had. You’ll open up the video express application in the browser within Chad GPT and it’s going to actually copy and paste it in the browser. This is experimental. I’ve been testing with it. It works sometimes where it doesn’t work sometimes. But it’s a good way for you to understand where the future is moving, right? As we head into the direction of complete automation of what we are trying to achieve. Okay. So cool. So let it do it. Let it let it do its thing. Okay. I also want to show you real time that it does take time. You don’t want to rush these things because it’s doing the whole process for us. It’s uh doing the visuals and all these other things and it’s getting us the thing. And you see also it’s doing some research as well to come up with some accurate information uh on trends and stuff like that. It’s up to you. It’s optional. You want the fast one, it does the fast one, it does the small one, the small. Okay. Now the idea is when you go here the cool thing is that on this particular page the video express/ uh workflow page as you can see here um I’ll show you in this particular page if you scroll down to the bottom there’s a copy link you can copy and paste this or you can build with GPT there’s two options why do we have a copy link as well so copy link works because what happened is TGPT moved into a desktop application recently and that desktop application allows you to kind of launch these GPTs from the desktop application itself because you know natively it’s moving more into the work side of things where it’ll do the task for you similar to what claude has been doing but it’s much cheaper to do it with this compared to anthropic. Anthropic is a lot more expensive and um it’s kind of prohibitively expensive really. I mean I’ve done the stuff I’m showing you right now here it was costing me like $50 uh worth of credits and stuff like that in anthropic which is ridiculous. Here it’s kind of like uh here as well included now. Uh so as this is generating the other thing I wanted to mention as well is you can even use the free option. So for example you can let’s say copy this link I can open up a new chat window. Let’s open that up here. So it’s opened up. So let’s say here I remove the pro right I can cut this pro and then I can uh switch the uh options as well right what I wanted to kind of use. So like let’s say for example you wanted to just uh do this. So we can say turn um like we can say like create uh 60second explainer on how Amazon delivers packages within one day let’s say now in this particular case it’s going to be much faster you see this so this is what I was saying you can do both options either it’s going to be the fast option or the slow option sometimes the slow option if it’s you need more accuracy and without many edits you can do that okay so now you see here um what I kind of wanted to explain to you is you see how it says uh copy this text to video prompt with native narration. So it says landscape and all that. So you can uh do this. So if we copy this and we go back into video express. So pay close attention here. There is an option under create with AI. Okay. This option right here. You’re going to click on this and then you’re going to have text to video. Right? This option. And under text to video, you’re going to go and all you’re going to do is click on the advanced option because it’s going to give us the prompts. And then you’re just going to go in there, right click and paste. Okay? And then here, uh, you can click on the advanced mode here. You want to enhance or whatever. It’s up to you. It doesn’t really matter. Share to public. If you want to do it or not, it’s up to you. The idea is that when you check this advanced mode option again, it basically does dynamic uh length. So even though it says 8 seconds sometimes it might come 9 seconds right or it might come 7 seconds. So instead of like abruptly cutting or like uh not not extending by clicking on this it it automatically dynamically does it which is useful for this feature. And then you click create video and then make sure that whatever the format is you have it selected. You see landscape is selected landscape is selected right. So that’s the approach if you will. So you got both options basically is what I’m trying to say. So all right now you come back here and then you can come back to scene two and then you do the same thing. Come back here and then again while that first one was being generated you can come back here paste this again. This is scene number two. Okay click create and then while that is being generated you you can come back here and then uh you can do the next one. Okay depending on your account type in video express uh you can generate like parallel videos as well. Um with the all access account you can do a lot of comparable videos. So while one is being generated you can do the other one. If you have the basic account you can do only one at a time. Uh so since it’s a onetime payment with no credits or monthly fees or anything like that you have a lot of freedom to upgrade into that all access account. Anyways um once you have this workflow anyway so once you clicked on this this is the third sequence and then you come back here and you do the fourth sequence right and then um then let’s do that delete. do this, create pretty straightforward, right? Copy paste, copy paste. Uh, pretty straightforward stuff. Now, why these videos do well is because people want educational videos and educational videos generally get a higher what’s called uh CPM, meaning advertisers are willing to pay more for educational videos than they are for entertainment videos. So, in the previous session, I talked about educational videos and entertaining videos, right? Entertaining videos are good as shorts where you want subscribers to get monetized on YouTube channel. But when you want like let’s say you want targeted buyers to watch your videos or advertisers to spend money on your videos. Let’s say if AdSense is your goal, you want to do that. Okay. And come back here, scroll down. Let’s uh get to this and then right click copy. Create this one. Pretty straightforward, right? Nothing really fancy here. So this is the seventh scene. And then we’re going to come back and see do this, right? Okay. Now, while this is being generated, I also want to go back into the other sequencing where it was doing all that thinking. Remember it did these sequences uh for us right what I can do here this is the beauty of chptd that’s why we’re doing these workflips with chptd um is you can say turn this into 9 by6 shorts video okay so you can do this as well um and uh then it’s going to do that reworking and it’s going to have that intelligence now right so that it it knew what it did in the past and it’s going to basically effectively you see this right it’s going to do this right here Like so you see what I’m saying? So instead of 16×9 now you turn into 9×6 a vertical video. So let’s do that also so that we see both use cases and you’ll see if there’s any mistakes being done in the in the in the workflow so that you know that’s the point of these live streams. It’s not to make it exciting but to see show you what can happen because what with AI due to hallucination even if it’s the best video creation model or most expensive thing out there when you try to create it there is some hallucination. So sometimes, you know, one or two clips might not come good. So you got to go back and edit it. No matter how much money you spend, it’s just the way it is right now. Which is good because it’s going to give a human uh more ability to add uh bring value to the table, right? Effectively, if you think about it, makes more sense if you think about it, right? So uh that is basically the uh goal here. Okay? So that when we come in here, we’re going to change this and then uh it’s going to do that 9×6 thing right now. And it’s going to generate that 60-second thing. And again the idea is the same. What changed here is that for 9 by6 there’s some copyrightiting hooks right we want to make the interest first five seconds interesting and the scene transitions are going to be a lot more uh dramatic. So all that stuff are fed into the chat GPT thing to do that. So when you want it as a vertical the same video it’s going to update that for you. And again you can do that with a free account as well. Sometimes with three accounts, what happens is that that context, right, it the chativity just starts hallucinating a lot more. I don’t know why that’s the case. Maybe they want you to pay or whatever. It doesn’t matter. But it’s my goal to give you the best outcome, right? It’s up to you what account you prefer to use. I don’t really it doesn’t really matter at the end of the day, right? You may have to prompt it a little bit more with a free thing to get it the way you want it. But at the end of the day, it’s the same thing. All right. So, uh, so here you see this, we have these quality checks that I’ve done. This is what I’ve taught it. Okay. So, um, as you can see here, what I’ve done here is narration is complete and naturally paced. One consistent narrator across all scenes. Each prompt is self-contained. One continuous shot per scene, one camera movement per scene, no dependence on previous scenes, meaning that every scene is unique. And this is very important for these um, animation style videos. And stable geometry and lighting, minimal visual clutter, and clear explanatory story arc. Right? These are the things I fed it uh to ensure that every video comes out properly aligned and video comes out well, right? Obviously, so um so even when it does a different style of video, it keeps these uh quality checks in place. So when the video is done, it’s checking that too on the other direction. Okay? So you don’t have to do much checking on your own. So anyway, so while this is uh getting generated, you can do the same thing for this as well, right? In the faster mode. in the faster mode. Again, I’m just going to show you because of the live stream scenario how it’ll appear. See this? It’s going to update it and it’s going to show us the faster one. See vertical and then on and on and on, right? It’s going to give us these little copy and paste toggles under uh it to for us to work with. Okay, it’s pretty straightforward, right? Nothing really fancy there. All right. Now, the other thing if you come back into the application, let’s just bring this uh these uh videos together. So, they’ve generated so far all of these. So if you notice this, these are all in realistic format. Did you notice that? So uh that is another option for you to do and also you can do like walk style videos as well. So that’s something that you want to keep in mind depending on what you’re trying to do. So in the case of like uh uh in these free accounts sometimes it will do these uh kind of things where it might do a realistic style and you want it in 3D style. Okay? So keep this in mind. So when that happens you’re going to prompt it to do it more in the box style, right? And that is the reason why I say sometimes paying for that uh whatever $8 a month or $20 a month is good for the intelligence. But anyway, let’s just take a look at this thing.
So the audio got placed into the mix as well which I’ll play with uh play to you in a bit so that you know what happens across right. So let’s just uh upload um you know these as well. Let me show you so you understand what I did here. Okay. and how you can uh basically fix some issues that might appear as well in the end. Okay, so let me show you. So one of the things was it’s more realistic in nature not cartoon right so that when that happens right when the text to video basis that happens you’re going to have to make sure that you tell it to do uh walk styles so sometimes these things they don’t clarify clearly right so you’re going to mention that and I’ll show you that in just a second but let me just render this out and show you as well how this will appear in real time okay so I’ll say realistic um narrative video okay and then let’s just export this I’ll play this to you so you can actually visually see how it’s going to appear. Now, when errors like this occur, it generally occurs because of that like I told you, right? The free account uh type thing. This is what I was trying to tell you in the previous training session. It’s worth paying that 20 bucks or eight bucks or whatever it comes to because think of it this way, right? A video used to cost me $500 or so per video and it’s 20 bucks. I mean, you can do unlimited relatively with my workflow, of course. So, that’s the way you want to think about this. And you can do the same thing with clot as well. It’s just that as I said uh before Chad GPD is little bit well in fact substantially cheaper uh than claw and it’s much faster to for a lot of these things. Okay. So anyway so let me show you now uh in this particular case where I have that u more that plus account thing which did it. So let me show you the vertical video set up while that thing was generating and um let me show you the difference. You’ll understand what I mean. So the real speed trick. So let’s go back here and this is the reason why I kind of tell you why these live streams are important. A lot of these YouTubers, they’ll they’ll show you, oh yeah, I did this copy paste and you do it and then this type of thing occurs and you never get to replicate the results. The difference is in what model you’re applying. What actually is it the free account? Is it a paid account? Is it because of uh like a special prompting thing you missed all these matter, okay, these days. Okay. Now, anyway, let’s just save this project. Um, and I’ll say realistic Amazon video, let’s say. And then, uh, let’s open up a new one. And in this particular case, I’m going to pick vertical video. Remember I showed you? So, let’s just pick the vertical option. And then what we’re going to do is do the same thing. And then I’m going to go with create with AI here on the right. And then I’m going to click on um text to video. And then now I’m going to go to advanced again. And right click and paste this. Okay. Now, what’s different here? Okay. Check this out. What happened here? It says modern editorial coll uh collage animation design for a shorts video. Okay. You see it did that. So keep this in mind. Okay, these little details do matter ironically enough. Okay, and um you can just um create this and then I’ll show you the difference between this and also what happens when you just do advanced mode as well. Okay, let’s do both of these so you understand uh what’s happening behind the scenes. Okay, I’ll remove you can automatically enhance the prompt as well. It’s not a big deal. So click create. So what automatically enhanced prompt will do is sometimes these things are gibberish, right? One continuous 8-second shot vertical 9×6. This is not important, okay? because it’s a given right obviously it’s a given that it’s going to generate this in this format which I GP sometime sometimes ends up adding in for no reason at all okay so uh you see that a neutral North American accent says this so this is basically where the narration will happen within the uh video itself okay so that’s the basic idea so now let me show uh this to you so if we go in uh this realistic narrative video I’ll show you the logic behind it okay so let’s just save this and I’m going to upload this in so you can see it uh how this video appears which what I showed you right one of the things is because I use the free account and this is what’s going to happen and this is what’s probably happening for a lot of people out there which I noticed as well when we’re doing it with beta testers and stuff is they’ll use the free account and they won’t be able to replicate the same thing that we show uh many of the times and so anyway let’s play this one you’ll see what I mean by voice over in the video okay that’s what we did in video express in the new millions of packages reach customers in just one day that speed starts long before you place an
Popular products are stored in fulfillment centers close to where customers live, reducing the distance every package travels.
After you click buy now, software selects the best fulfillment center and immediately sends picking instructions throughout the building. Packages travel through sorting centers before moving by vans, trucks, airplanes, or regional delivery stations toward their destination. Fast delivery is the result of forecasting, automation, transportation, and careful planning, working together from warehouse to your doorstep. Local delivery drivers receive optimized routes, helping them deliver many packages efficiently throughout neighborhoods every single day.
All right, pretty cool, right? So now this is the reason why we want to actually test things properly because often times to what we see to what happens it often times does not do a good job of it. Okay, these are the things with uh these things you want to basically keep in mind. All right, so now let me share my screen here real quick. I’ll show you something else so you understand because this is why we do these live streams so you understand learn what’s happening how AI actually works what are the uh implications of you’re not doing it properly and stuff like that okay so remember I showed you so if you come here let me play this okay you’ll understand
so this is the video that came now notice something right when I did this vertical 9×6 shot it still did in this format Why did it do that? Well, it did it because of my error. Okay, user error. So, go back here in text to video. What I did is even though the prompt was correct, I should have chosen vertical 9×6 which I didn’t do. Okay, these are the things you got to understand. So, often times even though the prompt is right, the setting is wrong in the application. Ain’t going to help you. Okay, so anyway, what I’m going to show you is you see how the prompting structure changed the video style now and but I got the style I wanted. Right. So, let’s check this one out here real quick. So, I’ll play this video so you can see it as well. Let me just download this real quick. Um, so you can see it so that you understand why it’s so powerful to have the voice within the video like we do within video express. Then I’ll show you how you can systemize this whole thing. Watch.
One day delivery starts long before you click buy. The real trick is distance. Keep likely purchases already nearby. You see that? So what happened is in the prompt itself. Okay, let me play this again. Um, in the prompt itself, the voice is existing because I prompted it to speak this and Chad GPT gave me that as well with the workflow that I gave you, which if you join late, you can get it for free here. You’re going to need video express. You’re going to need Chad GPT to use this. All right, so in that link for free. All right. So, you see that? So, let me play one more time. So, you see Amazon’s one day delivery starts long before you click buy. The real trick is distance. Keep likely purchases already nearby. Do you see that? So what was the difference? The difference was in the free account it doesn’t think even though I gave it all the instructions. In the plus account it tinks. It’s only difference actually. All right. So you get the point, right? So now let me just do this whole workflow now and I’ll show you how this workflow would work on a vertical video. Cool. And then you understand this process um in a nutshell and then we’ll walk you through and I’ll explain to you that as well. And I’ll show you another cool trick you can do to have video express you take over the browser um and be able to kind of do this automatically. Okay. So let me show you this uh as well in the application now so you understand this and you can um play around with it so that it gives you a better understanding of these things. Okay. Um because it’s all within the application. So you want to make sure that you understand the logic behind it. If you don’t understand the logic even if it does it automatically you kind of miss the point, right? and all the updates we make, we keep this logic in mind when we do these these workflows I’m building for you and all that stuff. Keeping this logic in mind. All right, let’s go back to um my GPT account. So here we go. Okay, this is um right. So let me just copy this. So let me just see what I’m showing you. See, copy this in video text to video. It’s saying right for native narration. So what I’m going to do is I’m going to go one by one each scene and I’m going to copy this. I’m going to go click create uh from prompt create with AI text to video. Okay. And um let’s click here. Click advanced. And then now remember I told you vertical video. We want to make sure. Okay. Click vertical. Right click. Paste. And what I try to do is um in the advanced mode I do tend to try to um enhance my prompt as well. Sometimes it depends. Okay. There’s no hard or fast rule. You can leave it, you can do it. It’s up to you. Okay. Uh the idea is though keep in mind that if your prompt is way too big sometimes it may not come as well. So that’s why I’m always trying to check. So if you check like let me explain you’ll not be able to hear it but when I checked it I did two videos remember. So here this one. Okay let me highlight this so you can see it. This one was where I did not enhance the prompt and this one I enhanced the prompt. Okay so basically right so let’s play this.
It’s almost the same. It didn’t change at all in my opinion. Just this one had a little bit better animation when I enhanced it. So it all depends. So let’s do that because in this particular case it did do so. So let’s place this prompt again. And I’ll select advanced mode. Okay. And then what I’m going to do is I’m going to and keep this selected automatically enhance my video prompt. Okay. It’s all depending on your testing because every style is different. Landscape is different from vertical is different. So I’m selecting vertical here. It’s quite different across the board, right? So anyway, let’s just create this and then let’s go to the next one and let’s do this process step by step. Copy and then um right click paste create and then copy. Come back. What I’m doing is control C or command C, command V. Okay. Uh basically if you’re using the Mac, right? Pretty straightforward. So come back here, create, and then um come back here, create. You I think I believe you can do up to uh five at once or so uh if you have the all access account at once since by the time you copy paste the one or two videos should be able to be finished. It’s so fast in video express. Um so that’s pretty cool. So click create and then um well let’s come downstairs down here. Let’s go to scene six. Copy scene six. Uh copy paste and then come back here. All the settings are saved, right? So it’s easy to copy paste basically. Okay. And notice one thing unlike the create video from prompt option right here. Create video from prompt option where consistent character and all that stuff is important. We don’t have to really select 3D style or 2D style or realistic style and anything. The prompt itself, this section does all the work for us. Okay, very important. So you understand? All right. So is that clear now? Okay, you’re following along. So now what we did so far is the vertical video. See? So first one is being generated here as you can see pretty fast. So now when you go back here to media uh library. So you see all the videos are being generated here. So now what I can do right is uh I can come in and I can drag and drop these into the timeline like so. You see now it came as vertical. Did you notice the error was before 16×9 cuz I selected incorrectly. So now let’s drag all these things to the timeline one by one. Um, and then let’s um put these up here like so. And then um uh let me show you something cool here. And then this way. And then these these um like these some of these are also pending. So 1 2 3 4 five of these are pending. So um let’s pull them as well here uh as they get created. And the other thing I did here is not all of them will be uniform. Okay, very important. Like meaning there’s going to be some scenes are going to be 3 seconds long, some will be 6 seconds long and on and on. Why is that the case? To make it interesting. Okay, it’s so as for a human being, a user who’s going to watch it, a person, they want it interesting, right? They don’t want it to be like boring and uh that sort of thing, right? So, uh that’s the other reason why we do what we do. Okay, in order to ensure uh this is kind of exciting through and through. Okay, cool. So, now let’s drag and drop this one. Uh you can actually go and do this where you can go grab a cup of coffee or you can go for a walk or something and it’s fine too. I mean, up to you or you can go check out something and it doesn’t really matter. It does this uh pretty fast. Anyways, okay. So, this is the last video pending in the workflow. Once I merge them all together, I’m going to give it to you. Now, you’ll notice something. Even though I said 60-second video, okay, even though I said 60-second video, it’s already in the timeline here just past 1 minute. Okay, that is okay, right? Because what I did as a system I told this um uh thing to hey even if it goes higher it is okay because I want it to be interesting. So you can do cropping right you can come and crop it to make it under 1 minute if you want to by just uh using this crop feature but the idea is make it as exciting as possible. Okay. So, since I did it in a rush, maybe one or two scenes might be on here and there, but the idea is that let’s export this thing. And then we’ll say um vertical Amazon video, right? And then um I’ll show you exact. And then you click U there. Cool. So now it’s it’s basically being processed. So the logic here that we are trying to do is depending on the format we want to make sure we prompted that way and we basically go backwards and u do this process. The audio is native to the workflow. Okay. How we manage to get consistency on the text to video is because in the prompt itself we say neutral American voice or whatever. So the the workflow will create that. So that way because it’s AI generated it can maintain it throughout. Now, if you want your own voice, then you can go to create with AI and then you’re going to go to create video from prompt. Okay? And then you’re going to click here and then you see where it says narration video. Okay? In here, you can link it with clone voice.ai and other applications of ours, right? And then you can basically create consistent voice, consistent character videos with your voice cuz it has to the cloning part. But if you just want AI generated voice where it’s going to do a radio broadcaster voice or it’s going to do let’s say a specific style, you could do the text to video option with the with the image to video aspect here. It gives you a little bit more control with consistent character and stuff. And in that case, basically this narration video is a better bet to be able to do that if you want your voice. If you don’t want to use your voice, well, it it’s not needed, right? You just follow what I just explained to you. Uh just extra steps obviously, right? So anyway, so now the video, let’s see if is it done? Is it done? Cool. So now this is done. Let’s just take a look how this looks like. Again, I did it in real time, so it could not be perfect also. But it’s good idea for us to know if we made any mistakes, right? So that’s very important. All right. So let’s just see how this video looks like. All right. So let’s play this thing.
One day delivery starts long before you click buy. The real trick is distance. Keep likely purchases already nearby. Amazon’s one-day delivery starts long before you click buy. The real trick is distance. Keep likely purchases already nearby. Amazon’s one-day delivery starts long before you click buy. The real trick is distance. Keep likely purchases already nearby. Forecasting systems predict what each region may need, then place popular inventory in fulfillment centers close to local customers. When an order arrives, software selects a nearby facility. Robots move storage pods while associates pick, scan, and pack the item. It travels through the middle mile by truck and sometimes aircraft toward a delivery station near your neighborhood. The labeled package joins conveyors and scanners that sort it by destination, then send it toward the correct outbound trailer. At the delivery station, packages are grouped by driver route. Routing systems organize stops using locations, traffic, and weather patterns. A driver follows the optimized sequence to your door. One day delivery works because inventory, people, machines, and timing move as one coordinated system. Pretty cool, right? Did you notice one interesting thing there? Because of me dragging and dropping it very quickly, one of the sequences had that southern accent, but otherwise it’s very neutralized accent. And in the prompting to make it exciting, it’s showing male and female voice. So if you want, you can go back. I’ll show you. Okay, pretty cool. So you can say in here basically it’s the same logic. We can say something like I’ll show you in this uh thing to make sure that the voice is consistent, right? We can say make uh uh change the prompts to I’ll show you in my screen one more time. You can’t see my screen obviously. So let me share my screen. So here so you can say make the prompts in the GPT I created for you. Okay. Uh which was in the workflow page, right? So this one this one right here. So in the GPT that you see you can come back here and you can change the prompts to ensure the voice is consistent. um example neutral American male voice right so we can even put like 25 25 year old um uh uh we can say something like um excited trailer like right and then you do it so what it’s going to is going to change all the prompting structures again for you to copy and paste and keeps that consistent or you can move it around at the end of the day you want to make it as exciting as possible. Right? So that’s the basic idea of why we do the GPT along CH GPT effectively to kind of be able to edit these things to come back and paste it. Now the other cool thing that you could do, okay, which uh I’ve been experimenting with as time goes on, I’m going to kind of walk you through this stuff uh more. Okay, obviously is um Chad GPT created a new feature when you download it, right? The desktop application uh called browser. Okay. So, kind of let me show you how it works to kind of understand it a little bit better. And I’ll actually open up one of my um my Macs here so that you can kind of see cuz I have my other computer. I’ll show you how it works in real time, especially on a copy paste uh basis. Okay. So, how this works is you can go to Google right now and I’ll show you. So, you can go to Google and let me show you this. Actually, let me go to Google and you’re going to go and say chat GBD desktop app. Okay, this is going to be on Mac, Windows, whatever. I’ll show you in desktop, in the Windows, and I’ll also show you my Mac because some of you might have a Mac, some of you have Windows. Where does it appear regardless? Okay, the reason why is I don’t want to be left out here in terms of the compatibility stuff. Okay, so you’re just going to go download this Mac OS or Windows uh depending on which uh uh device you’re using. And then once you download it, let me show you the Windows thing right now so you can see it. Uh actually, let me open this up here. One sec. All right. So, in the Windows thing, you can come I’ll show you. Um, it’s going to appear like this. Okay. And like this, right? It’s going to be like this. It’s a blank project. It’s going to ask you for project and all that. Leave it for now. It’s going to also say uh approval. You can say approve for me is the manual option. So, you can say approve for me uh instead of full access. But if you’re a desktop, I’d rather you pick approve for me. Uh so, it’ll do it. You can pick whatever model you want to do. doesn’t really matter uh at this point in time. So if you toggle your side panel on the top right here, okay, in the top like this part right here on the top, you see it’s going to say browser right here. So this is a new feature they released just I think couple of days ago and that’s why I wanted to be in the forefront of AI, right? If you click on this, it’s going to allow you to open up anything. So let’s say VideoExpress.ai, right? You can open up Video Express like so. Okay. Now, what you can do here is you can say something like Okay. And this is what uh I’m kind of training this thing to do with uh this is still in the works. Let’s say not perfect. So, what we can do, let me show you. We can go back and um so let’s say like so, right? And um let me show you. Okay, let me see here. Okay, it generated this to me, right? for example, the scene thing right here, this vertical video. So, what you can say is come here just um in this chat. Let me actually show this a little bit better for you. All right. So, in the chat, okay, and I’m going to change myself slightly so you can clearly see this better. All right, cool. So, in the chat, you’re going to say something like this. You’re going to say
you can’t still see it, right? So, let me remove myself and do it. All right. Cool. So, here you’re going to say something like this. Okay. Open up um the video express app in the browser and then open text to video option and paste this prompt under advanced um uh selection and then uh and hit generate. Okay. And then you’re just going to right click and copy and paste this. Okay. There we go. And you can test this. Okay. See what it’s going to do on the right. Okay. Uh, pretty cool. All right. So, let me uh bring myself here a little bit. My tiny version of me. Okay. Let’s say actually this one. All right. So, what is doing is I’m using the inab browser control because this task depends on interacting with the video express page that’s already open. I’ll navigate to the text video. place the prompt in advance and start generation. Okay, there we go. You see to the right, it opened up the screen the right over there. Okay, are you following along? So, this is where automation will be done as time goes on. And because of the fact that we’re not charging any credits or any uh monthly fees and stuff like that with the uh V Express, you can kind of use this to kind of make it do stuff for you and effectively. And you can even do where you can make it do multiple things for you. See there what it’s doing there um on its own. Okay. Right now this feature is only available for the plus accounts uh with a with a browser control thing. Um so it’s a good idea for you to upgrade it just to check it out. Uh but as time goes on I’m sure they’re going to release it for the free accounts as well. But you see this. So it generated the video for me on the fly. Okay. So it’s pretty pretty cool, right? And now what you can do on top of that is check this one out. Okay. And um it’s pretty cool. So this is what I’m saying uh which is even better is you can share this um let’s actually do this one right here. You can go to the workflows that I have for you and you can share this uh conversation if you want or even the workflow itself. Okay. And you can give it to go one by one. Okay. So for example what you could do is uh for instance you can say something like uh give Chad GPT browser um an option browser the instructions to launch uh these one by um let’s basically launch the browser
uh and uh and Open uh video express app text to video mode and advanced mode checked with 9×6 vertical option selected. Right. And uh do all the seams text to video prompts one by one automatically. Okay. uh give me the uh it’s basically like give give me the format to easily copy and paste right so you can do this in the GPT so right now one of the problems is that uh if you have a GPT to basically load the GPT in the um account is not that easy because it’s more web web based style right so it gives you the style here it’s going to take a couple of seconds uh to basically do it then you can just paste that file right it’s going to be a markdown what’s called markdown Okay, which is like a txt text file, right? Copy and paste it. It’s going to do that. It’s still work in progress because um you know it’s it’s kind of like finicky but it’s good to know where the future is headed correct on this uh direction as the browser control is uh going into this u fashion right now. And the other advantage with this is that um for example is that as we also modify video express into newer versions uh this will become even more powerful right because uh now we know how AI has vision now and it is doing actions on the screen. We now can kind of you know
make the application a little bit more compatible with this which our guys and our team gals in our team are working on. So it’s going to be pretty powerful as time goes on. Is that exciting or what? Right. So it’s doing that now. Right. So uh while it’s doing it, I’ll show you another uh pretty cool thing that uh um is being done here. Just to give me a sec while I load up. So this is basically done in the Windows uh in the Mac 2. I have a Mac app. I’ll actually show you right now so you can see it. Um let me open that up. One sec here. Um yeah. So all right cool. So let me show you in the Mac app and I’m I’m I’m going to show you uh a virtual window that I have in Mac obviously. So let’s open this up. So this is my Mac basically. Um you see this same logic. Okay. And it’s going to uh first step it’s going to say import data from your Chrome if you want to have any Chrome in your passwords or uh passwords or something. And then my Mac is the same logic. Right. So uh the idea is that it only works for desktops in particular and I asked it would this work for free chat accounts. It’s like no not the full automated workflow. Free accounts can try only limited tasks right built-in automation is currently excluded from the free and the go plans. That’s what it told me but anyway it’s going to get to this eventually even in the free plans right so what it did is opened up the browser in my Mac and again it pasted this thing and it generated an image for me automatically and stuff like that. So as we go into this direction, it’s going to be pretty powerful, right? If you can imagine. So I did the same thing. I open up the browser in my Mac uh right here that you can see. And then I opened up uh app the video express. I logged into the app and then I told it go open this stuff and do this for me. Correct? So you see it from now from a systems thinking perspective how you can now uh make these AI tools we say like CHP whatever to take over your browser and the GPTs I’m building for you you can make it do the tasks right and everybody’s computer is different everybody’s style is different so browser is way better than if it is going to be in any other way in my opinion because sometimes you know with APIs and some of these things it might get very complicated for you uh as you start calling it and you need some coding knowledge and all that kind of stuff at that point. This is more easier for us all to understand. So anyway, so let me go back here and um so now that’s exactly what I’m trying to tell it to do. So I’m telling it to give instructions in a way in this GPT for uh the actual uh desktop application here to understand and one by one paste each of these. So in other words, all it’s doing is where let’s say if it was um you know if you go up here, let’s say it was um you know scene number two here. So we go here, we copy this. All it is telling it to do is uh do the same process for this prompt, right? You see how I’m asking it right now and I’m pasting this prompt. That’s exactly what I’m telling it to do automatically basically. Okay, does that make sense? So now watch what happens. Okay, so uh up here um it’s basically going to pay close attention to here. Okay, so right here as you can see. So, and and it’s basically telling me right now I’m using the same inapp browser and um and um browser workflow again. I’ll reopen and use the advanced thing and I’m doing the same thing is what it’s telling me. Okay, effectively. So, you don’t need a lot of intelligence for it. It’s just you’re telling it to kind of go do this for me, right? And keep continue doing this for me. Okay, now this is me asking it again and again. What I’m telling it to do in the other window, right, the GPT is simplify this further, right? Uh you see that it just created the video and it did the video for me now. Um and it’s basically kind of finished. As you can see here, it’s done here. It says, but let’s go and see if the video is done. H yeah, cool. See, video is done. See this right here. So this um this this video right here. Okay, I play it. So this is the video I generated so far. Okay, you see so you can do it basically with the desktop. That’s the cool thing as we get into this direction, right? And there’s another video that’s being generated here on the right um effectively. Okay, so uh if I go back here uh let me see if our Yeah, see this this is what we like to call an instruction, right? So um like check this out. Task automatically create seven authorization. I authorize you to do this blah blah blah and on and on right so copy this just this little uh copy option this one on the top that’s why I said give it to me easy copy and paste option so you can copy it and this is what I was trying to say it’s experimental because sometimes it works sometimes it doesn’t but it’s good idea to know where the future of video creation is headed right it’s not video editing tools but rather these things where these AI uh tools or agents they like to call it will do the work for us okay so if I open up my uh brow uh my uh application here and then I’ll again change myself slightly a little bit so you can see the screen better. Um let me see here. Pretty interesting. So let me do it this way. Yeah. So let me just paste this whole instruction format. It’s going to say showing the text field or whatever doesn’t really matter. So there we go. We’ve done it. Click on uh send. Okay. And then it’s doing the thinking now, right? And it’s what is it saying here? I’ll read the attach task instructions first. Then carry out the requested video express workflow in the app browser. I’m using the browser control skill because this requires repeated interaction with the live video express UI right and then now it’s going to basically take those prompts and one by one uh basically go in and do it right uh in the application does that make sense so that is the general idea of what we are trying to do one by one it doesn’t do it like a human that’s the problem so sometimes it’s no it’s going to be very finicky on how it does it see now it opened up the text to video so let me uh put myself on the screen right now so you can see it uh better. So, well, let me do this this way. Right. So, you see this right now what it did um it opened up went to advanced tab right on the right and I I don’t have it open. See like my hands area here I’m not doing anything. So, uh basically it’s uh uh it’s kind of going into the function now and kind of copy and pasting it one by one. It’s going to take some time but the logic is it’s going to do everything um on its own on the desktop uh as well right and that’s what the browser automation is. So as I build more uh automations and stuff, we can improve it. But the thing is even the application side, we are kind of working on some cool things to kind of simplify this a little bit. But you do see where the future is headed, right? Like if you think about it. So I have my hands right here. See? And that’s hiding anything. So just so you know that like uh it’s just doing everything on its own, right? So uh basically this is where the future is headed to. The idea though which I’m trying to um drive home now is that as we know these things can do very rudimentary copy paste type of jobs uh very rudimentary like you know go here do this type of task right very straightforward stuff but the thinking part still comes from us mainly in a sense that if you’re not satisfied with something uh you have to go and tell it to fix it because the end person who’s actually uh watching it uh is basically going to be a human. It’s not an AI uh person, right? So, just keep this in mind um as you uh play around with it, right? And sometimes it’s going to do some of these things on its own. So, just be careful. Okay? Now, here you see this what um I wanted to show you is that in order to prevent abuse of this thing, we do put like for you to wait until other videos can be made. So, watch out for that as well. uh because of the problem is like a lot of people will come in and then like thousands of videos like this it just doesn’t make sense. So what we are trying to do as well is we’ll probably allow this uh to be happening where we change the pricing structure a little bit for people. If you want this level of automation down the line uh we’re still thinking how to price this because uh it’s it’s just not practical for us where thousands of videos can be made every second by user. We don’t have many GPU servers um as well. We’re not like an open AI uh kind of company and so it’ll be something where we’re still thinking on how to go about doing it but you see where the future is headed right on this uh direction and where the computer will be taken over and you know a lot of the cool things can be done on the browser end um and uh that’s why we are kind of like thinking how to go go and do it in such a way where you know everybody benefits on it but at the same time it makes it economically viable for us because of our pricing um structure and all that stuff. it doesn’t make sense for us to do it without some type of u pricing structure that makes sense. So if you’re interested in this type of thing do let us know to be a beta tester so we can track this thing uh stand goes on from this direction and make it viable across from us and yours. So the last thing we want is like uh it takes hours to generate these right which is not practical. So do let us know uh if you want to be a beta tester for this stuff which we’re working with GPD and openAI uh people for more integration stuff like this uh for our applications. So it’s pretty cool though right if you think about it. So for the logic though you see how we started with a manual process and now it worked really well then we kind of went in a little bit deeper and then we uh did the uh direction of like the automation here where it’s generating all these videos you see here on the on my right on its own. You see? So the logic started for you to understand what video works well which the workflow does and then bring it to the automation side of things. Right? So that’s the basic idea and I just want to be very clear with you because I don’t want to uh be um there. If you look at every other application in the marketplace, one of the main things they all do is they charge credits, they charge tokens, they charge monthly fees on top, they charge uh extra throughput where like if you want parallel generations, it costs you more money, you have to pay extra and stuff like that. Uh the reason why that’s the case is that GPU servers, these Nvidia servers are like millions of dollars right now. Even to buy eight of them right now cost like uh just recently we got a code which for $300,000 and one person uses up all eight if it’s eight generations. Okay. So let’s say for example you go in in the browser and it’s clicking clicking clicking that’s eight uh of these million dollar servers, right? So meaning eight in one is a million dollars or so that it costs. um just one server cost about 50,000 when you link it together into eight can go all the way once installed in the data center with electricity and all that cost so much. This is the reason why Chad GPT closed down Sora, right? Uh their video platform cuz it’s not viable for them economically even though they have billions, hundreds of billions of dollars in venture capital and stuff, right? So, uh when you’re interested in beta tester, let us know. Email us in the support email. We’ll uh because we want to figure out how to price this thing properly as time goes on because it won’t be included um the ability to generate thousands of videos obviously, right? So, some type of a system in place where we may have to charge credits or something like that down the line as future heads into this direction. Okay, we’re still thinking. We’re not sure. Uh so do let us know if you want to be a beta tester on this on that email that I posted up there. But it’s a good idea to understand this right as we head into this direction. Um and that’s the reason why you you keep hearing where somebody in YouTube says, “Hey, you know, I created a video. Hands up. I’m an influencer. I did it like where you see dance or something and I created like a 5-minute video.” And then you ask him how many tokens, how much how much in credits it cost them. And they’ll say it cost them $500. Cost them $300, right? And so this is the reason why for us to make it uh affordable but at the same time practical uh we want to make sure that first of all the quality is maintained and second of all it’s done in such a way where it is uh helpful right for you as well and it makes it economically viable for us. So you don’t we don’t disappear like 2 months later cuz it keeps crashing our servers and stuff uh cuz one person is using up like thousand people’s worth of uh usage let’s say right uh it doesn’t make any sense. So anyway that but this is a good way for you to understand the future. Okay. So if you see the practical uh sense of this it generates us all the videos. Now you see on the right it’s doing all the stuff. Now all you got to do is you can have it even merge it into the timeline and create the video if you want to or you can drag and drop it yourself manually. Now to put it in perspective if I come back here. This is my actual other browser. Okay. Um if you see let me show you this browser. Um so here if you see under media library here all these videos are being made so you’ll see what I’m talking about uh what is being done in that other browser within Chad GPT. Okay, you see these videos, it’s Chad videos building it and you’ll notice it. Um, if I open up here, see this video is finished. You see, just let me show you. Highlight it. So, this video is finished. Then, um, if you come back here. Okay. Uh, oops. Let me pull this up. Come back here. This video is finished here. You see? Because it’s the same account that I kind of did it with. So, that’s the basic idea as well. Okay. Pretty clear, right? So, that’s the basic idea of how we are trying to look at the automation side of things. But you got to understand though or the logic of it uh from a step-by-step standpoint. Okay. So, uh that’s the general idea. Okay, does that make sense? So, now you can get an understanding of this. So, in this uh this one, what I would suggest uh we did this option, I want to show you this. So, because characters don’t mean anything. What we are really asking for, I want to talk about here. Um so, let me kind of explain this a little bit better. So, characters don’t mean anything. What you have to really do is dynamically update it. So what we did even in the create with AI right where there is um lip sync HD option here okay um and even in the case of uh video uh and audio prompt if you followed my other trainings you’ll know we created something called the advanced mode what this will do okay is you see how we have like 12 uh 10 seconds or something like that limit now what we do and this is quite critical for you to understand is that it dynamically updates It’s the length. Okay. Sometimes, you know, some voices are too fast. Some voices, let’s say Australian accent or British accent might be a little bit slower than the American accent, right? So, even though it might be characters, it means nothing, if the length dynamically changes based on the speech uh speed. Correct? So, the better way to do it is regardless of what you put in the prompt, it’ll dynamically update it. Same thing applies even what I just explained to you where if you go into the advanced mode, you select the advanced mode, okay? and you go in here and then um you you have this right here. You see? And then you pick on this option. Okay, this advanced mode option. I’ll show you this. And then click on it. But do not turn on this. Okay, do not turn on this manual. Do not. Okay, if you just turn on this, you can enhance the problem if you want to or don’t want it. Doesn’t matter. But if it’s just the advanced mode is checked, we have a mechanism where it dynamically updates. So instead of characters, it focuses more on the length. Okay, very important because what happens is that I could have 115 characters like you suggested and then what will happen is in that scene an American accent will be let’s say 9 seconds long Australian accent will be 11 seconds long and so on and so forth and it will not fit that length here right so dynamically adjust it for you which is what you want okay so make sure you use this mode rather than looking at character lengths and stuff because the scene has to make match right has to make sense has to be good quality So that’s the goal. Okay. And um and u uh this is the other thing that’s thank you so much for Christine here. Right. The main thing I’m trying to explain uh folks is what’s happening across right now is that um the technology is moving so fast and these companies if you will if you will these companies like Chad GPT and these other folks are like literally subsidizing their plans, right? Like meaning they’re losing money on these accounts. So you might as well leverage it, right? these browser tools and stuff like that. We don’t have the capability cuz it’s not like we go in and banks are like JP Morgan and stuff are lending us billions of dollars right and so if you look at it us or seed dance and cling and some of these platforms which do video and audio and stuff 11 labs is another one uh they all don’t have that capability although they raise a lot of money too but they don’t have liked kind of capability so uh ability so what’s happening is that we can subsidize it effectively we still want you to use the state-of-the-art the highest generation stuff but at the same time. We want to make sure that it’s viable for us uh so it doesn’t you know uh it doesn’t become a situation where uh we disappear right because it’s not viable or it slows down your generation which is the worst part even worse than that uh where you want like some person comes and then they’re like just for fun of it oh I’m just going to generate thousand videos in a minute yeah we do have thousand servers maybe but that for that one person we spend let’s say $10 million worth of uh expense on that it doesn’t make any sense right cuz to have thousand servers that’s the kind of money you got to spend and in the power of it, right? Even if you’re a genius in terms of optimizing, which we do, um because every video goes through one server for us, just for reference, okay? Each server cost about $40,000 right now. You don’t believe me? Hey, let’s ask our buddy Chad GPT right here. Okay? Uh because a lot of the times people don’t understand and they came uh they get uh frustrated. All right, so let’s just do this instant, right? I’ll just ask it. Instant whatever. Uh how much does an Nvidia uh B200 is the service we use? um server cost uh each and how much does a rack of AEX um servers cost, right? And it’ll tell you right now. Okay, there we go. So, you see this like eight uh servers, right? Which is what we kind of have to use because that’s the minimum they want us to buy. They don’t sell them individually. Uh it cost $500,000 just to buy them. Now to install them in the data center and the you know like the uh uh server racks and then all that all in cost you got to double it right so uh what’s the total cost of 8x uh B200 GPUs fully installed in a data center okay because it’s very important people understand because sometimes people well I want everything for free well it doesn’t work like that in real life I wish it was um so you see this right uh like about $700,000 give or take And then on top of that uh you’re looking at you know electricity and stuff right so all in node 800k you add other things right power of the rack and all that you see it’s like a million dollars for like all in like 800k at least now that’s that it’s only eight renders at a time you understand so let’s say we want to buy um how much would um 100 of these cost allin cost right in a data center uh plus monthly costs for posting. Okay, this is the level we had to spend these days. You understand? So when you talk about quality, you don’t fool around here. You see this 45 to 52 million. Okay, and this infinity band and this and that install hard or 50 million right here. Okay. And uh it requires at least a 1.3 megawatt what facility of power and uh that’s going to cost at the bare minimum like 200 grand a month, $200,000. Okay, this is the reason why OpenAI cancelled Sora. Okay, um they kind of said you’re not going to pay this kind of money for video stuff. It’s easier to just do text. Okay, it’s going to be less intensive. Um and it’s easier to subsidize for it as well. Okay, just understand because uh we are trying to think about it like where we want to give you results, the outcomes and we want to um you know basically give you the best of the best stuff but also at the same time this is the level of stuff we’re spending investing behind the scenes right tens of millions of dollars behind the scenes. Uh right now we have a mix of the older generation, some of the newer generation, we have some optimizations in uh on top of that and all that. But if one user comes and they do 100 videos, it’s 100 servers launching, right? Each whether it’s done in 30 seconds, 1 minute is irrelevant, right? So this is the reason why everybody moved into credit system, monthly system and stuff like that. And credits on top of monthly, on top of throughput like let’s say you want to do five parallel generations, you got to spend more, right? You want to do more than one, you’re going to pay more. like when you go into say uh you know some of these chat GPT accounts and stuff and you try to generate one image it will not allow you to sec generate a second image until the first image creates right it takes you two to three minutes let’s say this is done on design by purpose okay for this reason okay so yeah anyway to kind of summarize this video uh so I I want to make it short and sweet and stuff like that what we discuss right now is uh if you will is this workflow that I created for you where you can come in and go into this page page uh which is this vdexpress.ai/workflow and then um you can effectively go to this page and then up here if you scroll down you’re going to see this um right here uh part which is going to be this motion graphics animation workflow. Okay. And just make sure if you’re doing it in um in text to video style that you force JB to generate it in that 3D style or paper cutout style or collage style whatever it is that you want to do. Okay? because motion graphics means many different things and that’s why I made it a little bit more generic so you can kind of guide it a little bit to get what you want. Okay. And then um you’re going to come and you’re going to click on this uh build uh with GPT option right here on that page. And then you’re going to come into the GPT. Okay. And then um you’re going to open up like this and you can work with it. And of course of course as I mentioned before it’s going to generate us uh the whole flow. It’s going to do this one continuous shot and whatever whatever in prompting. Okay. And it’s going to just basically do 9×6 or 16×9 that you can just copy and paste. You’re going to be easy copy and paste options like so for you to copy and paste into GBT. Now in here you’ll notice something I have like the like the pro account selected or plus account. You can just use the plus account is good enough. The free account I just want to be clear is not as intelligent. So it may not be that easy for you to replicate what I’m trying to show you. So just keep this in mind as technology is improving. Uh it’s a good idea for you to upgrade at some point. Uh pay for a month or two, 20 bucks. Like it’s better than Netflix, right? At least you get some value out of it to build videos. And what would have cost $800 to create one video with a video editor? You can do that now literally with video express and uh this tool uh for like peanuts, right? For nothing. Um so basically that’s the idea. Now the second thing I I also explained to you as a part of the workflow is that um as time is going on um as you can see in my uh Mac as well as my um uh Windows computer here it’s going into this browser task right so as time goes on let me actually pull this up here so you can open up your browsers and you can basically do the automation by opening up uh appwexpress.ai AI right like what we have here okay and then it can basically uh do each of these scenes uh on its own and it’s going to generate and it’s doing that one by one by one okay and you can go to bed can do whatever let it do its thing okay um right now video express does have some parallel generation limits so keep that in mind I think three to five videos depending on um your account type if it’s the basic account it’s probably one video at a time but you can tell it wait until one video finishes before you start another video in the chat here okay so it’ll do that Okay, as an example, so intelligence is being done this way. One of the things that we have been focusing on internally as well, which I kind of wanted to explain right now is that as automation goes into this direction, uh a lot of the uh stuff is when we say work, okay, if I write this down here, say work, okay, it is actually basically what drives results for you. Okay. And so what we have to do internally in my own end which I’m planning on doing over the coming weeks and stuff is test right test across my other YouTube accounts, Instagram accounts and stuff like that. Uh we get you know in a year billions of views let’s say and if we do that what is working what is not working right now and then only what’s working we create an automation like what is shown here. Correct? So we automate. So if I basically put it as a different uh font here. So we automate what works. Okay. It’s not just about telling uh Chad GPD build a video. It’s about telling it build this type of video in this format and don’t do this do this. That’s my job. Okay. I’ll my team and I we have like 12 people now, right? 12 people. All we do is all day long test these things across. This is working this is not working. Uh what can we do really well? What can’t we do really well? uh vertical video for this style, horizontal videos with style and we’re building these workflows for you. Okay? Uh with time, it’s going to be even better and better with time and getting improved upon. But just keep in mind that the future is headed where we are going into the work side of things now. Okay? It’s not just about research. It’s not just about asking it to write some scripts. It’s about telling it, hey, open up the application and do this work for me. So now two things are important. Okay? The application that you use better be good, which is video express in this case. The second thing is the support you get from people like me is also good where they’re on top of um the stuff right we’re really really on the cutting edge of technology and the third thing which is probably the most important thing is that it is affordable okay and that’s the thing that takes us a long time to figure out how do we make it affordable for everyday people like yourself and um not like a corporate enterprise like for example it’s easy if let’s say your employer is paying for an account right then uh they can pay for the $200 a month account or they can pay for like the seed dance 2 which costs like $4 for like one video which is like 15 seconds long, 10 seconds long or something, right? And then you got to test it out. Not all videos are good. So every video you create costs more money. Easy. But when you’re a person who wants to try stuff before you believe in it that it’s actually generating you revenue or traffic, leads or sales, then that has to be a situation where the person uh in this case our job is to make it accessible to everybody, right? And so that’s what we’re working on right now internally. We’re doing a lot of tests. We’re working with beta testers. If you ever want to be one of our beta testers, let us know in the email I sent you. Support Express. Really happy to uh uh see how to push it to the limits, so to speak, and uh structure this stuff. Uh but thing is even here this is the thing you got to still build what’s called a system correct okay even if it can do all the thinking the system is very important okay the system the automation as we like to call it has to be very very u to the point uh straightforward and it has to basically work with a structure right so that’s what we’re planning on doing as well okay so uh your homework for Okay, let me give you some homework. Not the best fun, eh? Uh, you didn’t you didn’t expect that, did you? Okay, but no, let’s let’s put some homework here. So, it’s uh actually useful. So, your homework is very simple. Okay. Um, so let’s do that. Your homework will be if I open this up um is to get the desktop application for GPT. Okay, that’s your homework. So, so let’s say uh you’re going to get the desktop app. Okay. Okay. Desktop uh child GPT account. Uh if you’re familiar with cloud, you can kind of do this too. But understand that with their latest models like that Opus 4.8 and like um their Fable 5 or whatever. So expensive you can’t do anything with it before it runs out of uh uh that plan even if you spend a lot of money with it. So you get the desktop app. You open up the browser tab. Okay. The browser is going to be on the top here on the right like so. You open up the browser and you are going to log into your video express account by going to app.videexpress.ai and then you’re going to use my GPT elements, right? You can um basically tell it to you know do one by one or whatever the case is and you can do initially copy and paste but after you get the flow and you get a hang of it, you can basically make it do all of it. Okay? And just keep in mind that there’s going to be parallel generation limits. Keep that in mind. I want to be transparent. So there will be parallel uh generation limits. Um I believe it’s about 3 to five um depending on the uh all access account type you have. Uh if it’s a basic account, you’re going to do only one at a time. So basic uh account equals one video at a time. And I believe it’s uh like 3x to 5x that on the on the uh unlimited/all access. Okay. So you just tell this to Chad GPT. It’s like okay wait uh I think it’s three to five depending on I believe the type like text to video is a little bit more than image to that kind of thing. But anyway you’re going to tell Chad GPT that hey this is the limit. So wait until um one finishes um before uh before you do more. Okay? Does that make sense? That’s why I’m telling it before you begin the other one. Okay, it’s going to do that. It’s going to wait and then it’s going to the next thing and the next thing and the next thing. And as I said, one of the things we’re working on with our beta testers, you can let us know if you want to be one support at video.ai that I left the link on is let me actually put it up actually. This is the link uh this email you let us know is that um what we’re working on is if you really want it automated where you just sleep and it does everything um basically for you but at the same time has that editability component and stuff. We’re working on how to make it viable uh because of the cost of these things. Electricity costs have gone up with the data centers like 30%. Uh RAM and CPU memory and hardware everything has gone up in the last 6 months uh by up to 600%. So meaning let’s say was $300 before it’s like $1,200 $1,300. You must have seen it, right? So it’s impacting everybody not just us. It’s impacting Apple with their iPhones and their MacBooks. They raised prices on that. It’s impacting your cars. It’s impacting everybody. Okay. Uh but the good thing is that we have you in mind. We kept you in mind here to make it affordable. So as time goes on, right, you can be rest assured we are on the cutting edge of these things uh more than anybody else in the industry. Okay? Cuz we’re using this inside and out. We have our own machine learning AI team and all that. So we’re doing all these tests as time goes on to improve the systems and literally push it to the limits. Okay? So that’s your homework. Get the desktop app. It’s going to be here. Going to go to Chad GPT. You’re going to desktop, right? And in the desktop app, you’re going to have two options. I want to show you these two options. If you pull it up here, let me pull this up. Um, you’re going to have this codeex or or chat GPT option. You can use any one of them other codecs. So, like if you come back here, you’re going to be either work or chat, either or up to you, right? So, um, you can do work in my opinion. So, this was what you want to select when you want to do this kind of stuff with browser automation and stuff. Is that clear? uh because the chat is slowly dying out in chat GBD as time goes on. Okay. So yeah. Uh can you plug this in Gemini as well? No. The problem with Gemini is you can’t do this browser stuff right now. And it’s also the other problem with Google seems to be and I want to be transparent because we do this stuff every day. Uh Google has what’s called a TPU. Okay. Uh this is what’s called a tensor processing unit or in simpler terms what that means is their own hardware. Okay. Own hardware. So the problem with Google is they don’t want to use Nvidia’s GPUs or Nvidia service or cutting edge. They want to use their own stuff. So because of that they’re super slow uh to release stuff like this browser stuff and all that. You’d have to wait 3 months, 4 months, 6 months and stuff for Gemini. I I see your point. I’m just trying to tell you that is like the issue with Google is this. So they want to make sure that they save costs and they want to put it in their own hardware. So because of that um the main headache comes because of that logic which is they want to use their own hardware. So because they want to use their own hardware but take 6 months 7 months or whatever to come. Okay. And um the other uh thing as well that I wanted to mention is uh um this this is a good question which is uh say uh I I want to do a video for a law firm 60cond video that answers questions. How do I go about doing it? Shpt, right? You’re going to say make a Q&A style video. Okay, question pops up on the screen and uh you know the answer comes across. In that case though, you don’t even need um uh like video express if it’s just the basic stuff like that, you can probably even just get chat to do an animation style uh video where it’s a Q&A type thing, it can do it. Uh video express is more advanced stuff it can do, you know, can do stuff with animation, people, lip sync, all these other stuff. So it’s way beyond that uh uh thing. The other structure you could do is um you can do it where um person one is talking person two is talking lip sync style video uh one after the other. So in that particular case what I would do is I would go into video express and then I’ll go into uh it can be either or uh let’s say text to video for the sake of argument for speed of explaining this for you. Uh but you see it says actor one script actor two script. So you can say actor one is saying asking a question and then actor two is answering the question. Right? So you’re going to basically click actor two and then this would be the question and then this would be the answer. Okay. So uh basically I can say question whatever how do you uh let’s Gemini question right? Can I use this with Google Gemini right? Um let’s say um can we use it with Gemini is the question and the answer could be not at this time as their uh developments are slow right so that’s the answer you get the point and you can describe the actors uh here okay the actor is this this and the actor two is this this this okay that will be the idea of uh if you want to do it in that a question and answer format okay and um the other uh uh point as well that I want to mention is as we are heading towards u AI doing the work okay what you really need to understand is you need people who understand systems thinking okay and this is what my focus has been uh over the last year or so that’s why I kind of disappear and come back once in a while with these trainings so with systems thinking what are we what we’re trying to do is we’re trying to get AI I in this case Chad or Claude or whatever Gemini was another one which being asked whatever it is we’re trying to get AI to understand a repeatable system. So we want to make it repeatable. Okay. And not every task is repeatable. For example, if I was to trying to create uh let’s say a video for um YouTube and in that particular case I just say create a YouTube video. Well, it can, but then there’s a thumbnail requirement. Then there’s a title. Then there’s description. Uh there’s also in that YouTube video one intros and uh to make it the first 30 seconds interesting. So there’s so many components behind it. So what happens is a video creation is not uh for example one system. It’s not one system. It is tons of many many mini systems. Correct? Subsystems like thumbnail would be one system. title would be one system, right? Uh if I say thumbnail, uh then it’s going to be let’s say long form shorts, right? So shorts is one system and then uh you’re going to have one for the long form content, one is 3D narration style video like we discussed today and then one will be realistic style video tons. Okay, so the trick here is that the the repeatable system is not one system but many systems. Okay. So the job of ours is to test which system is good, which system is there and figure out the subsystems. Which is why if you look at it, I said go and get this workflow. And we go to that page really. Uh let’s go to that page here. You’re going to see a ton of workflows, right? So you’re going to see like you see this here. We say pick the workflow that fixes the bottleneck for you. Uh motion graphics idea to 3D documentary. um 10,000 hours into AI video animation automation like copyrightiting and all that and then we say uh first frame last frame here and then we say 3D animation short film and then we say realistic talking video you see what every one of them is a different subsystem right it cannot be one because AI doesn’t understand right if you don’t stress enough like hey once the video is finished go and give me a download link uh for me to download it doesn’t even know where to give that download link. Should it give it to you such that you can download in your computer? Should it upload it to a Google Drive? Should it email you? It doesn’t even know until you specify. So, a system has to be end to end accounting for it. Like you’re teaching a baby what to do. Okay. Okay. Baby, now go or like let’s say an 8-year-old. Okay. 8-year-old, now go uh to the store. Now go to the second refrigerator on that convenience store. Then you’re going to find in that place you’re going to find um you know let’s say orange juice and it’s going to be two options for orange juices. Then pick the second one which has a little bit of logo with um you know green logos which says I don’t know like uh Del Monty’s orange juice or something and then that instruction has to be step by step. You see if you just say go get orange juice it doesn’t know which orange juice. Where is the orange juice? You see same logic applies here. Uh so the person doing it has to know has to be an expert to be a really really good expert in that stuff and has already built something has already gotten results doing it. It’s not just dabbling with it and then uh repeatably creates that and the only way we found out is each of it becomes its own workflow. Uh and it’s only going to get worse with time because as these things are becoming more intelligent they’re also becoming more stupid in a sense that they become more broad and then you want a specific use case. it starts thinking all kinds of weird uh directions and it wastes a lot of money as well in terms of usage tokens and in terms of time and everything as well. So that’s the beauty for being a part of e express because you get to you know have us do all the trial and error so you can save time, money, resources, energy, everything. Okay, and that’s the other beauty as well. Okay, so do the homework, finish it up, be ready because the future is very exciting here. And as with all feature updates we’re making, we’re going to be moving into this direction where you know uh clone voice cloning will be automated. Uh video express video creation will be automated and also the other part will be the behindthe-scenes logic of what needs to be in a good video in order to get it to get results. Whether it’s an ad campaign video, whether it is entertaining video, it’s an educational video, whatever the use case, each will have its own different elements within it that uh chip will never be good at. Never. It’s impossible. uh because it’s trying to think like a like math 1 plus 1 equals 2 in real life you know when a YouTube video could have everything bad but it was in the right place at the right time leveraging a trend like the FIFA World Cup right now and it might get millions of views for no reason other than that right that trend that’s happening now and these things are not smart enough to understand those things so uh that’s why human is very valuable in the mix okay so that’s the basic idea all right so okay then um this should be the training right now. So, watch out uh for our tutorials page. We’ll upload this replay and also I’ll be creating shorter forms of this and I’ll be sending emails out to you as well. So, watch out your email uh so it’s simpler, easier for you to um add up. These are more Q&A style so they go a little bit longer, but you will see shorter videos as well in the tutorials page of Video Express as well as you’re going to see the shorter form in the emails that we send out to you and stuff. So, watch out for our emails, okay? It’ll come from Video Express or myself, Paul Honer, or whoever it doesn’t matter. You’ll see it, okay? Um uh these things as well. So that said, thank you for being here. Have a great weekend and I hope you enjoy this. Hope you understood the logic behind it and we want to make you the masters of this stuff. Then you want to you spend that time to learn it from me here which is valuable and that’s that’s appreciated as well. Okay. So if you enjoyed this after but I didn’t answer any questions for you, post it in the chat box as well. Okay? Uh meaning chat meaning the comment section of YouTube. I check every single comment in the YouTube channel and not only because I it’s interesting for me to reply but also because I want to improve any systems we are building okay the workflow improve uh video express improve the presentation itself the training live training itself and stuff like that so after this is done it becomes a replay on YouTube live um then you can post in the comment section as well any questions that I couldn’t answer today or you want to get answered in detail you can post 10 questions it’s okay no I’ll answer every single one of them okay it’s a way for me to improve our stuff, right? Our services, our products and stuff as well. Okay, so thanks again. Wish you all the best and I’ll talk to you very, very soon.