Free AI Video Creation: Step-by-Step Guide & Transcript

[00:02] learn how to make long-form videos without hesitating. And the timing is almost perfect. A few months ago, this market felt more saturated, [music] but YouTube has changed some rules to push back against low-value uploads, and that

[00:15] projects. What those people didn't understand is that YouTube doesn't care if you use AI to create content. What YouTube cares about is whether that content gives value to the viewer. It has to be useful, entertaining, or teach

[00:28] something. So, in this video, I'm going to show you how to create videos longer than 10 minutes in any language and any category using only completely free and unlimited AI tools. But better tools could come out tomorrow. So, if you

[00:40] don't want to miss them, subscribe and leave a like. First, we need a text AI. and we're going to add this first prompt. Here, you can fill in a specific video you want to make. And if you don't have a base idea yet, you can leave it

[00:54] When you send it, the AI will research which topics have the most potential right now. In other words, topics that are getting views, but do not have too much competition yet, so we can create AI videos around them. It will give you

[01:06] the one that gets your attention the most. Then, save this full text in the next step. I'll leave all the links for the tools I'm using today, plus free text and image AI tools you can use on my website in the

[01:19] to find everything at the end of the video. Now, we're going to add this >> [music] >> In these boxes, first write how long you section, paste the selection we copied before. The AI will give you several

[01:32] video ideas inside the topic we chose, and these are ideas that have a better chance of working. Now, tell it which idea you liked the most and how long you will give you this full structure divided into blocks. Each block includes

[01:45] a narration line of around 5 seconds, an image prompt, and a prompt to turn that so we can follow the order without getting lost. The The step is turning this into video clips using the narration line above each block with

[01:59] free AI tools. The prompt has already made sure that the story is not only original and eye-catching, but also entertaining for the viewer. That is what helps us avoid uploading the same low-value AI content everyone else is

[02:11] making. First, we need tools to generate images and videos. I researched several one problem. The watermark was too annoying for a serious video. But inside that same platform, I found a way to fix it and generate the clips cleanly. Let's

[02:25] go into Meta AI. Inside this platform, we can do almost the entire process, but we need to follow the right order so we don't run into the problem from before. Start with a new chat, click create image, and now go back to the text AI

[02:38] from the beginning. Open the first block, copy the image prompt, return to Meta AI, and paste it here. When you send it, my favorite part is that Meta AI gives you four good quality variations with enough variety to choose

[02:50] into video. We'll leave the video part for later. Now go to block number two, copy the next image prompt, and send it again. Keep doing this, one block after another, until all the images are ready. Don't worry because this tool is

[03:04] completely free and unlimited. There is no section to buy subscriptions or pay you want your video to be, you'll need to generate more images, but you can Just be patient until you have everything. Once all the images are

[03:17] everything. Once all the images are ready, we need to turn them into video. >> Good background videos are only half of the work. After this, we still need to >> [music] >> and so it doesn't look like the

[03:31] low-value content everyone else is creating. Let's go step by step. To turn these images into video, I thought we were going to need another tool, but right now, free AI video generators are in a rough spot. Most of them are either

[03:43] starting to move behind paid plans, like Rock, or their limits are too tight, like Google Flow, which can work for a few individual clips, but is not useful for bigger projects. So, the option that remains is Meta AI.

[03:56] But, as you already saw, it has that annoying moving watermark problem. Luckily, I found a solution inside Meta without using external platforms or illegal methods. Go to the first image we generated for block one and return to

[04:09] the text AI. In block one, under the image prompt [music] we just used, there is also a video prompt. Copy that video prompt. Then, inside Meta AI, click the image you like the most out of the four versions. Scroll a little and click

[04:22] custom animate. This panel will open and here you paste the video prompt. When you send it, a few seconds later, the image becomes a 5-second video clip with clean motion. Now, repeat the same process. Go back to the text AI, [music]

[04:36] copy the second video prompt, go to the second image, and turn it into video. You can do what I did. First, generate all the images and then turn all of them into video. Or, you can generate each image and animate it right after. The

[04:48] videos in the same order you're going to download them. But, if we download them like this, we get the same problem from before. To fix that, first finish generating all the videos. If one of them comes out wrong, just generate it

[05:01] again. This part is completely unlimited, so use that advantage. At this point, the clips are 5 seconds long and they do not have audio. We're going to solve both of those problems later. But, for everyone who wants to generate

[05:14] much higher-quality videos that already come with audio, I want to show you how come with audio, I want to show you how to create something like this.

[05:37] In this platform, you get the best AI tools for video, images, audio, and a lot more in one place without having to pay for several different subscriptions. that try to imitate them. I'm talking about GPT image 2, Seedance 2.0, which

[05:53] is currently the top video model in the rankings I checked, and many other tools that you can use inside the same platform. To create a video like the one we saw before, go first to the images section. I'm going to use GPT image 2,

[06:06] I get the perfect image to turn into video. The images are generated in 4K, and they do not have any kind of watermark. Once you have the image you like, click turn to video. This takes you into the video generator section,

[06:20] where we're going to choose Seedance 2.0, although there are many other models available, too. Upload the image again and add a prompt. Unlike most other video generators, these videos can be 15 seconds long, and they can also

[06:33] include scene changes, camera changes, and sound. And from the research I've done, Higgs Field is the most accessible place I found to use Seedance 2.0. If you want to try it yourself, I'll leave the link in the description and in the

[06:46] pinned comment. Thanks to Higgs Field for sponsoring this video. Now, if we you how to solve the download problem from before. Once all the videos are generated, click the one you want to download. Right click [music] and choose

[06:59] opens, look at this button on the left side of the menu. Click it, move the mouse over the video, and click again. Now, on the right side, this block will be selected. Double-click this part,

[07:11] copy it, paste it into a new tab, and [music] press enter. The video will open here. It plays without the error from before, and if you click the three dots, before, and if you click the three dots, [music] you can download it.

[07:26] keep generating for free and without limits. But, this is still not enough, because nice videos alone will not make someone stay and watch. We need someone to narrate them. And since I know many of you do not want to record your own

[07:38] voice, I have a very good solution. Go to the Pinokyo website. The Pinokyo we're using today will be on my website in the description, so you can get Once you're on the Pinocchio website, click download, choose your operating

[07:52] Pinocchio is a platform where you can install AI models and use them for free and without limits. Once the models are installed, you can even use them without an internet connection. When you're inside, go to the top right and click

[08:05] explore. [music] In the search bar, type Quinn 3-TTS. install [music] it. Just keep clicking through the steps and Pinocchio will inside, you'll have several useful options for creating the narrator. I

[08:19] about this model and several others inside Pinocchio, so I'll leave it in it after this one. [music] But the basic setup is this. In the model section, for the size, [music] always choose this one. Then click

[08:35] download. Do the same with the other two options that remain, base [music] and ready. Now you'll have different ways to generate your narrator. In voice design, you can write how you want the voice to sound. And if you go back to the text

[08:48] blocks, it already gave us a line this video best. Copy that line into voice description. Then in the text-to-speech box, [music] paste the narration lines from each block. When

[09:01] strong result. And if you don't like it, you can keep changing the voice looking for. >> A signal leaves a living world. It travels farther than any voice. Then somewhere

[09:15] something hears it. A civilization goes dark in seconds. the voice of someone you have permission to use and make that voice narrate any text you write here. Or inside custom voice, you'll have a

[09:28] list of preset voices you can test in case one of them fits the video better. >> With Quinn, there's no limit, so use it freely. Just keep in mind that if you use voice design, it does not generate very long audio files in one single

[09:40] narration in parts. Once we have all the voice-overs, go to the Suno website. Here, we're going to create AI songs to use in the background of our videos. Go to create in the menu on the left, and in this menu, choose

[09:54] simple. Now, go back to the text AI from the beginning. Right before the blocks, song description that fits the video we're creating. Copy that description, generate. When the song is ready, download it. Now, open any video editor.

[10:10] >> I'm going to use the free version of CapCut, which you can download easily. First, [music] add all the videos we generated in Meta AI in order, and match each one with the narration from the same block. The narration and the clips

[10:22] may not match perfectly in length, so adjust the speed of each video until it that block. When that is ready, add the song we just >> I also recommend checking the transitions and effects that do not have

[10:35] the diamond icon in the top menus, because those can be used for free, and they help the video feel more finished. And with that, the video is done. >> A signal leaves a living world. It travels farther than any voice.

[10:48] something hears it. A civilization goes dark in seconds. Maybe the universe is not empty. Maybe it is waiting. This is not a proven [music] truth. It is a terrifying possibility.

[11:03] The question begins with silence. Billions of worlds may exist. Some >> We created the whole thing using 100% free and unlimited tools. To get the prompts, the direct links, and access to the free AI tools, go to my website in

[11:16] the description. Open PDF guides, and you'll find the guides for the videos I have published so far. You can download them and access the information from and links without wasting time writing everything manually. You'll also find

[11:29] the free AI chat there. You can ask questions without paying, So, it is useful if you want a free chat option for prompts, questions, or quick tests. We also added a model that generates very high-quality images. You only need

[11:41] to sign in and start using [music] it. See you in the next one, and thanks for See you in the next one, and thanks for watching.