How I Automated a $100K YouTube Channel with n8n & AI (Faceless Method)

[00:02] million subscribers, but what's really impressive is that the creator doesn't show himself in these videos. All he does is create YouTube videos with children's stories, and each of these videos can rack up billions of views,

[00:15] not thousands, not millions, but billions of views. That's truly impressive. So, I wanted to do a bit of advanced research, so I went to Social Blade. And really, the annual revenue he can generate from this channel can reach

[00:29] impressive. And then I wanted to dig a little deeper And then I wanted to dig a little deeper with my research, so I found this PDF, sources from Social Blade. And here, I see that there are a huge

[00:43] number of YouTube channels that, for example, can earn thousands of dollars per month. These are YouTube channels made entirely for children's stories. I've done a lot a lot of research. I kept finding more and more channels that

[00:56] numbers. And I asked myself, is this business actually profitable or not? The idea is simple. I use this tool called N8N, an automation tool, and I asked it if I give you an image like this one. This is

[01:11] a picture drawn by my 7-year-old child. Can I automatically generate a video as a children's story with a script, narration, and have it automatically published on YouTube? So, instead of spending money and time creating videos,

[01:24] which takes a lot of time, can we automate it? And the answer is, of course, yes. I created a workflow, which I will obviously share with you for free in this video, and I'll even show you how it works. So, it's a workflow that

[01:37] can actually generate, starting from this image, what we call characters. So, this image, what we call characters. So, it will detect it will Mr.

[02:02] He heard the laughter of Curly and Blue Kitty playing downstairs. Let's go on an adventure. Curly and Blue Kitty were heading off with excitement.

[02:16] They soared high into the sky together and saw something magical. and saw something magical. A cloud of shining birds. The cloud of birds drifted away like cotton candy.

[02:30] The stars drifted away and smelled like strawberries. The best adventure ever, they said, hugging each other.

[02:50] workflow. I will show you how to deploy it and how to run it, how to set up all the parameters for free. Of course, if you actually like the content, know that the video will then be automatically published on your

[03:03] And of course, don't forget to support me with a subscription, because thanks to you, I'll try to create videos every day using automation tools. All right, let's get started by installing N8N, and then we'll begin building our workflow.

[03:17] So, the first thing to do is to download the workflow, and to make it easier to deploy later, I've divided this workflow into three parts. That way, we'll be able to understand what each part does. So, all you have to do is go to this

[03:30] link, which I'll of course put in the description, called nytn.arferus.eep. address, and you'll receive the complete workflow by email, ready to be imported.

[03:45] So, the source code will be shared with you for free via your email inbox. So, now, once we've downloaded the source code, of course, we need software to run this source code. And we're going to work with N8N. There

[03:58] are two ways to install N8N. Either you simply go to the official website and subscribe to N8N, but you should know that there are some limitations there. First of all, it costs $20 per month, and on top of that, you're limited to

[04:12] running only five executions at the same time. That means on this server, you can only have five active workflows, which is really very little for freelancers or entrepreneurs who want to run workflows on N8N servers, especially since our

[04:26] parts. So, three workflows is really you upgrade to a much more expensive plan. So, the alternative I suggest is description that will take you directly to this page, which is dedicated

[04:40] specifically to N8N. We're going to do this on Hostinger, because Hostinger offers a low-cost N8N server that's ready to run. If I scroll down a bit here, the system offers either KVM 1 or KVM 2. There are also KVM 4 and 8, but

[04:57] the two most commonly used are KVM 1 and 2. The difference is the number of processors. KVM 2 has two processors with 8 GB of RAM, which means it's really a powerful machine. If you're an entrepreneur or freelancer, I recommend

[05:12] the KVM 2, because you won't experience any slowness in creating or running your If I select here, I'll be taken directly to this interface. There's a small discount coupon that was actually offered by Hostinger on their official

[05:26] For those who are buying their first Hostinger server, you can actually use So, this is the code we're going to enter. So, go N8N. I hope it's still valid. We click on apply. So, this will give me, there you go, a

[05:40] 10% discount. So, actually, the recommendation is if you already have a server on N8N, you can create another server with a different email address. And so, as a result, with another email address, it's as if the system will see

[05:52] that it's a new user accessing. And so, we can use the coupon. There you go. We'll keep this little trick to yourself. So, now, down here, we won't the server as France. And another tip, here you have a free

[06:06] server with N8N that includes 100 workflows. confirm, it will actually give you the 100 most popular workflows on the internet, and it's really worth taking them, because they're free. So, you

[06:19] confirm, and then all that's left to do is click here to continue. Of course, Hostinger offers a 30-day satisfaction or your money-back guarantee. This is very useful for testing the workflow and seeing if the business really meets your

[06:32] expectations. If it works, you have Hostinger's 30-day guarantee to test and get started in the field of automation. So, now, what we're going to do is simply click on continue to access our

[06:44] N8N server, and we'll just import the workflows together. So, now, we're going to create the workflow. As you know, I'm on the interface, and this time we're actually going to import the three parts of our

[06:56] workflow. So, I click on create workflow, and here you have these three dots, which allow me to click to import the workflow from a folder. So, I click here on import from file, and here on my computer, I'll find the three workflows

[07:10] that I downloaded. So, by email, they will be compressed, zipped. You just need to unzip them to find the three files. The first thing I'm going to do click open. So, here, we've just downloaded the

[07:24] first workflow. So, automatically in N8N, it saves, so there's no need to click on publish or anything like that. I'm going to click here, click a second time to import the second workflow. Same thing, I click

[07:37] there to do import from file, and there you go, I select the second one, and again, I click open and same thing, I'll go back to download the third one. here to import, and now we have the workflow. So, if I come back here to the

[07:52] home page, I'll simply find the three files that I'm actually going to work with. We're going to try to explain each of these workflows, what their function is, and exactly what they do. And there you go. Now, I have the three

[08:06] workflows that have been imported. We're going to try to understand the first a little bit here. So, in the first part, we already have a first step, which is a step that will retrieve the image that was drawn by your child. For

[08:20] example, here is an image that my little one just drew. There is only one character inside. So, first, I need to scan it or take a photo of it with my phone, and then the system will analyze it and actually extract the characters.

[08:32] So, looking at the first part, here I actually put a small form that will allow, if I click on open. So, this is a form that lets you upload the image. I can already run it to see a bit how it works. There you go. It's a pop-up that

[08:46] will open, and then you simply have to upload your child's drawing image. file, and of course, it will convert it to file mode, because here, we're using artificial intelligence to analyze and understand the image. We can already

[09:02] take a look at this. I use Gemini, because it's a system that's very very strong in image analysis. Don't forget that Nono Banana actually belongs to Gemini, so Gemini is currently the best at analysis, much better than ChatGPT.

[09:14] related to prompts, everything that provides retrieval information. And then, the image analysis that follows is where it will actually give me the list of characters, if there are one, two, or three. The more characters you have, the

[09:29] more work there will be afterward. So, for testing, I recommend you put in one or two characters at most. If you put in a lot, obviously, just imagine, for each image of that character, so it's going to be slower, which is completely

[09:42] normal. So, that's the first step. It allows you to extract the characters. This step is simple. The second step is simply to create the characters. That's why here, I call on an agent, an AI that will use Claude, so Anthropic. For me,

[09:57] descriptions. Here, I use Claude Sonnet 4.5. There's also 4.6, but it's a bit more expensive and 4.5 is already more than enough to perfectly describe the character for me. That's why, of course,

[10:09] I use Nano Banana Nano Banana here. If I click here, I'll show you. So, of course, I use Gemini since it's part of Gemini. Nano Banana will actually create this character. And then, look, I'm going to save this

[10:22] character on this node here. This allows me to have 1 GB of free space on the That's where I actually save all the files I've created with the automation. So, there's no need to overload my Google Drive or my computer. Everything

[10:36] is actually sent to this node. The node is called Cloudinary. So, it's a website where you can create a free account. And of course, from that account, you can set up your API here, which will simply allow you to send and

[10:50] save all your images and videos on this system for free. So, here, at this stage, the system's task is to generate the characters. Then, once we've retrieved the characters, they will be saved online.

[11:03] And now we have the third step, which is an interesting one. This is where we're going to create the scenario. So, at this point, based on the characters I sent, the system will create the scenario. You'll see here,

[11:16] we'll try to take a look what I actually set is that each scenario should last the length if you want. So, for each sequence I said, give me a narration for 10 seconds. Once it finishes generating

[11:30] the narration, I'll simply have all the scenarios. And that's when I'll actually workflow here. start working with the sequences. So, now we've moved on to the second

[11:43] workflow. Where we find an important step. Um you could say that, in fact, it's the same step. So, first, of course, the system will retrieve the character because we'll actually need to get it

[11:56] Remember that here we actually created our character. And afterwards, what it will do, of course, is call on Nano Banana to be able to use scripts, scenarios, and narration to create scenes for this character. That's why

[12:10] here, the system will create with Nano Banana, we always use Nano Banana, and we will save each image, each scene we create here on our system. So, at the end of this workflow, we will have to call the third workflow. And that's

[12:24] where we'll talk about videos. The scenes will be transformed into videos, and so we arrive at the third workflow. Now, what this one will do first is help me generate what's called a prompt to create a video because starting from an

[12:38] image to create a video, we need a prompt. Why? Because this prompt will actually be sent to the Kidding technology, which is a very well-known technology today for video creation. For that, I will need what is called Atlas

[12:52] Cloud. Atlas Cloud is a website. You can come here and create an account. It gives you access to all the models, whether for images, you can even use Nano Banana with it. Personally, I use it much more for videos.

[13:05] are a huge number of models. There's Sedans, there's one, there's VO3.1. But the one I'm interested in is Kaling 3.0. This one is truly the number one

[13:17] for creating children's videos, and honestly, the quality of the production is extraordinary. So, you simply needed to create an account on Atlas Cloud and then retrieve the API from your back office.

[13:30] I just entered this API here. Actually, I use my own, but of course, you can create your own API and enter it here. What will happen? me. But the video alone isn't enough because

[13:43] with videos, we also need audio. We need someone to actually read out this system. So, what do I do? I use Eleven Labs. Labs. And here on Eleven Labs, there's

[13:55] something really interesting about the voice feature. You can actually add your own voice, like I did here. I added, for example, some voices. I even added my child's voice. So, I can use the voices that interest me, the ones I want people

[14:07] to hear in the video. In fact, we're going to use those voices, or we can simply use standard basic voices that are specifically made for videos. There Labs. So, here, my workflow will call on

[14:20] Eleven Labs, and each time, of course, it will save either the video or the audio to our storage space. And the loop, this is actually a loop that will run until all the audio files for my video project are finished because in

[14:33] the video, we have several sequences, and each sequence or scene has its own audio. Once that's done, I simply need to assemble, to merge, to start. We have to put the audio onto the video. And then, the small video sequences need to

[14:48] actually what this part does. So, here, I use this website. And by the way, with a free account, you can actually get free credits that can help you create hundreds of videos. So, what does this site do? It has two functions. The first

[15:03] site do? It has two functions. The first is to add audio to the video for free, and then to merge the videos since we have several sequences to produce a single video. This site is called Shot Stack. So, you can simply create an

[15:15] course, use your own because here inside, I'm using mine. So, now the system will give me the result. And the nice surprise is that I barely had to use Bloop date. So, what does Bloop date do? Thanks to this

[15:29] description. And the title and description are what will be generated here afterwards with this agent. I'm going to call YouTube and TikTok. And all of this is thanks to an API, to a node that exists here on

[15:43] N8N. You simply type in Bloop date. There it is. It's right here. By the way, on their website, you can create a free account. account. And you'll see that here in Bloop date,

[15:55] really interesting is that you can connect all your social networks. Once you create your account here and connect your social networks, you'll see that in the system, I can select the channel where I want to publish the

[16:08] video. I can actually set it to private first, so I don't have to make it public right away, allowing me to control it if I want. But all of this is part of the I want. But all of this is part of the system with the workflow from A to Z.