I've been researching hundreds of different AI video tools over the past few months, and in this video I'm going to break down the top tools for each category of AI filmmaking. We'll go through everything from image generators to voice generators to video generators and more. So let’s dive in and explore the latest frontiers in AI filmmaking together. Flux is the brand new AI art generator on the block, and it comes in three different flavors. The first is Schnell, which means fast, and this is a small and fast model that you can download and run locally on your own computer. The next two models are much more powerful and are called Dev and Pro.
The easiest way to use Flux is to run it on Replicate, a place where you can use cloud computing to run different models. Now you can use all three models through Replicate, and the costs vary. For the Dev model you get 33 images for a dollar, for the Pro model you get 18 images for a dollar, and for the Schnell model you get 333.
Now you can simply go to this page on Replicate and select the model that you would like to use. Let's take a look at the Pro model, and from here you just need to enter a text prompt. You can adjust the aspect ratio. The other exciting way to use Flux is through Grok 2, X's AI chatbot, and they've released an image generator in Grok that uses the Flux models. As you would expect from an Elon Musk product, these are fairly uncensored, and allow you to create images of Donald Trump with guns, George Bush with some white powder, and Mickey Mouse with a nice cigarette and a nice cold beer, among other things. Flux is therefore an affordable and powerful model that excels in creating realistic images.
Now next up is Midjourney, and Midjourney has been the market leader in AI image generation for the past few years , and in that regard it recently released a brand new update that gives you even more control over your AI art than ever before. images. This latest update is now a web editor, and it combines reframing, repainting, varying, panning and zooming into a single, unified interface. We think this makes editing your MJ images much more seamless than before, and is a huge step forward. The key thing here is that you can apply a number of specific editing techniques to your images in one place , and that includes repainting, which means taking a part of the image and changing it, and varying the image, which means touching up small elements. of your image, or pan, which means you expand your image up or down, and also zoom, which allows you to zoom out or zoom in on your image.
So let's take a closer look at how that works. For example, if we take this image, you can now go into the editor and from here you can adjust a few things. First of all, you can erase parts of the image. After you clear the selection, you can delete anything you don't like. For example, you might want to change this bird. You might also want to change the nose of this guy , and what we can do then is scale the image. So we can zoom out like this, and it starts to fill the frame, so we can make it this small . Now if you want to go back, simply press the back arrow and you can also adjust the prompt here, allowing you to update the image based on a new prompt. And when you are satisfied, simply press Send and you can generate your images that way. This is particularly useful for a number of reasons, when working with AI video one method is to take a starting frame and then define the ending frame, and then ask the AI video to animate between those two frames.
And now that we can better refine and control our footage in Midjourney, we can determine exactly how we want to animate different scenes . For example, I took this photo of a young lady, and then what I did was zoom out and change her facial expression so that the final frame looked like this. Now I can put both frames into an AI video generator and create a video. And this is a great method to get precise control over our work. And they recently released their latest version 6.1 , which takes the quality of their images to a new level.
Now the aesthetics that come out of Midjourney images are absolutely astonishing. And now they produce even more consistent arms, legs, hands and bodies, which have always been a challenge for AI to decipher. Image quality has increased tremendously and now we are at the level where it is almost impossible to distinguish between an AI art image and that of a photo. And now Midjourney is even better at displaying text within images, allowing us to fully unlock new possibilities. Today, Midjourney excels in a range of different styles, from graphic art to Gustav Klimt-style paintings. The cinematic style visuals are absolutely stunning.
Just look at the representation of these individual hair types. Now Midjourney is incredibly easy to use and has a great community around it. It also offers a pretty decent membership, with the basic plan starting at just $10 per month. What I especially like about Midjourney is the philosophy of the makers behind it. They have repeatedly stated that their approach is to create a tool that is useful to artists. They are much more interested in creating something useful and beautiful than in making money. And this really comes across in the way they interact with their customers. Midjourney CEO David Holtz has had a very interesting career and previously created this beautiful hardware project: a sensor that you put on your desk, and you could place your hands over it to control your machines.
And I was an early adopter of this and definitely enjoyed playing with it. That's why I have always followed the progress of David Holtz and his fantastic inventions. Now Midjourney is also planning to expand into other areas of AI development, and they are working on a 3D model generator, as well as their own AI video generator, which I am extremely excited about. Now the Midjourney team also say they are investigating how to create a storytelling tool.
So we have a lot to look forward to from the Midjourney team. But one of the biggest criticisms of Midjourney is its heavily censored nature. And there are many things you can't create in Midjourney. And of course this includes anything that is considered too gory or too violent. This includes detached body parts of humans or animals, cannibalism, bloody violence, images of someone being shot or bombed. Now you are also not allowed to create anything that is considered adult, mature we might say, adult. And they say you should avoid nudity, genitals, and fixation on such things . Don't fixate on those organs, and especially on fetishes.
And in some ways this will be one of the most challenging elements for Midjourney going forward, because there are other models that don't have the same level of censorship. So let's take a look at how Midjourney and Flux compare. Now, in my research I would say that both of these tools create excellent images with a high degree of realism. They vary a bit in how you turn them on, but the results can be absolutely mesmerizing with both tools. I'll show you an example. Here's an image from Midjourney of Will Smith eating spaghetti. Now this is the smallest model of Flux, the Schnell model, and you can say it is almost comparable. You would definitely say that MG takes it because of the more realistic way he uses his fork specifically. But Flux, Schnell are the cheapest and smallest models. Whereas if you look at the Dev model here , you can see that the quality improves dramatically.
And the same goes for the Pro model. And if you look at these two, you'd have to say that it's very much a matter of personal preference, that there's very little that's indistinguishable between the two. My experience is that Midjourney gives you a much more aesthetic output with limited prompt input, whereas with Flux you have to be more specific with your prompts to get anything aesthetic out of it . Now I want to mention a few other tools that are worth knowing for the image generation process. The first is Leonardo.ai. And interestingly enough, Leonardo was recently acquired by Canva.
So we expect them to invest a huge amount of money into expanding Leonardo's capabilities, and to incorporate some of that into the Canva tools as well. Now I like to think of Leonardo as somewhere between Midjourney and using your own local model, because it gives you much more control over the different elements of the algorithm. Now I would say that this sacrifices some of the streamlined nature of the UX, of the user experience of this tool. Now we'll look at how we can take these images and turn them into video. And this is an essential part of the filmmaking process because it gives us complete control over how the video starts. And there are some tools that also give us full control over how the video ends. Now I would say that Runway is the absolute best AI video tool out there right now. And this has some great features because it allows you to upload an image and easily create a video from it. Now you can also specify not only the first frame, but the last frame, which really allows us to guide and decide what the nature of the video is.
The biggest criticism of Runway is that it can be very expensive. To use the newest video model on Runway, Gen 3, it will cost you 10 credits per second of video. And with the basic subscription you get 625 credits per month. So you only get 60 seconds of video for $12, which is very expensive if you want to make something that will be much longer. However, they do have an unlimited plan, which means you can generate unlimited amounts of video for $76 per month. That's why I'd like to introduce you to a more affordable option. And this has some fantastic features too, as well as the ability to create incredibly lifelike AI videos. And this one is the Dream Machine from Lumalab. And what I especially like about them is their looping feature, which allows you to generate videos that loop continuously. This is great because then you can create clips that can actually be played multiple times. Now Luma is interesting because it has a really simple interface and you can only do a few different things.
So this is both a blessing and a curse. On one hand, it's very easy to understand that all you need to do is insert an image or text prompt, and those are pretty much the only features they have available. But on the other hand you start missing some really fine control or adding other kinds of features. But you can see from these videos that they are absolutely stunning and they give you a real sense of lifelike video. And I'm a big fan of Lumalab. Another honorable mention is definitely Kling, and he comes up with beautiful videos like this one. And I would say it's comparable to Runway quality. You can also use an image input with it and it has a very nice and useful website.
You may be experiencing problems logging in. It's been a bit temperamental for some people . And if this happens to you, you can try using the arrow keys on your keyboard. Now Kling offers you free credits every day, which I definitely admire as it keeps you coming back and using it for free. So if you only have small projects, this is a great option. However, if you want to upgrade, they offer a much more affordable plan than Runway. There is currently no unlimited subscription with Kling, but the videos are absolutely sensational. The last entry for image to video is Vidu, and there is something very special that I would like to explore about Vidu, and that is their character trait. And if you upload an image, you can ask it to be used as the character input , which is great for maintaining character consistency between videos. And Vidu also offers daily credits, which I think is very generous.
So my recommendation is that if you are just starting out , you can choose between the free credit options and just use those until you are ready to upgrade. This gives you the opportunity to test out the tools and see which one works best for your use case. Now if you want to go to the next level and really create the most professional videos, then I recommend you go for Runway and get the unlimited plan.
Let's take a moment to recognize today's sponsor, Vozo. Last week, Vozo won number one on Product of the Day on Product Hunt. Vozo is an AI video tool that lets you rewrite, re-dub, and lip-sync videos. Now you can do all this with simple text prompts. Vozo takes a video and fills it with completely new content. My name is AI Samson, commander of the AI video scouts, general of creative pioneers. So you can take a video and completely change the script and the voices and the lip syncing will be updated to match perfectly. So this gives users the control to fully customize the script voice and speed if needed, so you can use these new videos for your content or your marketing .
So let's take a look at how it works. What you need to do is come to Rewrite and Redub. From here you have the option to upload your own video or you can use one of the training videos. Let's take a look at this gladiator version. Hello, I'm Maximus. My name is Maximus Decimus Aurelius. Then you can go to Rewrite the video. And from here you can rewrite the video yourself, or you can ask the AI to do it for you. And to ask the AI to do it, all you have to do is go to this little glowing orb and say how you want it to be rewritten. There are also some prompt templates that you can use to turn it into an ad. So this is great for creating your own advertising products. So I'll make this an ad for Vozo. And then you can go ahead and press Apply to Script. Now go ahead and press Update speech. So it re-records this with AI. Now you can also change the voice by clicking on a line of text and going to change voice. From here there is a whole library of different voices that you can select.
Now the next step is to achieve lip sync. And what this does is it syncs the lip movements with the lyrics. So it looks much more realistic. And once that's fully rendered, you just need to go to the top right corner and export. And that's it. Now you have your own video. My name is Vozo, your ultimate video editing companion , commander of AI-driven tools, general of seamless video editing, loyal servant of creators and innovators. Who is Vozo for? Well, Vozo is perfect for marketers who want to rewrite and recopy content. Or for teachers who want to update old videos with new information. Content creators can also use this to create viral videos. Vozo offers three plans, but I'm most excited about their free plan because it gives you the ability to rewrite unlimitedly, reuse for 30 minutes , translate for three minutes, and lip-sync for 1.5 minutes.
And you can try all of this for free. They have an improved plan. One costs $19 per month. The other is $59 per month. And these offer enhanced features, rewriting and repurposing, increased translation limits, and also lip-syncing minutes. So check out Vozo today by clicking the link in the description below and see everything Vozo has to offer. Thanks to Vozo for sponsoring this part of the video. So now we have some moving images. It's time to add a voice to the process. And for that we can use a tool called 11labs. And this is my top pick for creating authentic, articulate and realistic voices with AI. The great thing about this is that it has some built-in voices that you can use to generate votes. Here's an example. Now you can also design your own voices with 11labs . You can use a text prompt to create a specific type of voice. This is a wonderful feature for designing character voices . The other thing you can do is you can clone voices with 11labs. And I cloned my own voice.
And so it sounds like this. This is my AI generated voice clone. And as you can hear, it's doing pretty well. And as you can see, it's almost imperceptible. You have the ability to design voices as well as clone voices and much more within 11labs, as you can also easily generate sound effects with text prompts. This is a great way to add atmosphere and a cinematic feel to your videos. You can simply type a short prompt and 11labs will generate it for you.
The great thing about 11labs is that they have a very generous free tier, which is best suited for all small and medium sized projects. So once you generate this, you'll be given four options. And as you can hear, they are very realistic. And of course these are perfect for use in your videos. So now you have voices and sound effects, but what we're missing is some music. So let's play that music maestro. And for that, we're going to talk about Suno, the number one AI music generator. And it's extremely easy to use. You need to come to suno.com. Now you get 10 free songs every day, which is a huge amount of free music, and you can click on them in a short prompt for creating your music.
So maybe I have a jazzy background for a cafe. What's also interesting about making music in Suno is that you can make music with lyrics or music with just instruments. So instruments are great when you're working on a cinematic piece. But if you're looking for something really quirky and full of life, adding some lyrics is also a lot of fun. This is my jazzy cafe. Nice. But one thing I like to do, this is a guilty pleasure, is make Disney-style songs , about myself or about other interesting things. And this is a Disney song about the AI Filmmaker Academy. Now the AI Filmmaker Academy is a place I created if you want to make entire films with AI. Here we really go through the whole process. I break it down step by step and we go through a whole project together, looking at everything from creating a story with AI to turning it into a beautiful AI movie.
And it's also a community where you can connect with other AI filmmakers, where we have live calls, lots of software discounts, and much more. So if you are interested in that, there is a special discount for you in the description below, check it out. Another notable mention for AI-powered music creation is Yudio. And this works in a similar way, but there is one feature that I particularly like on this platform, and that is the ability to upload a song and use it as a reference for creating more music. And this is great if there's something you like that's copyrighted, but you want something of the same ilk, with the same vibe and tone.
And to use that, you just need to go to upload when you look in the prompt bar . This is now a paid feature to upload the song reference, but their free tier is also very generous, giving you 10 songs per day. So let's also listen to a Disney-inspired musical about the AI Filmmaker Academy. Now that we have our voices and some music, you may want to add them to your video. And to do that we need to use lip sync tools, and for this there are a few options. Now Runway has its own built-in lip-syncing option, and it can create a version like this for you. Alternatively, there is also a tool called Pika which also offers lip syncing, and the Pika output looks like this. Now I would also recommend that another tool to explore is Hedra, and it will allow you to create more lip syncing. And as you can see, these are quite realistic, but it looks like there's some sort of airbrush approach being applied to these videos.
Another option is RenderNet, and they just released their own lip sync feature. So take a look at how this works. Now you might be thinking about how to upscale your videos so that you get the highest resolution possible, and my recommendation for this is tensorpix.ai. What I like about tensorpix is that they give you credits based on usage so you only pay for what you actually use, and it's pretty reasonable to upscale your videos. And what I especially like about it is that if we go to upscale a video, so let's select one here, so let's try this cinematic video, is that there are a number of other filters that you can apply that can help improve the quality of your video.
work. First of all, you have the ability to scale. Now you can also remove dust, so this is great if you're perhaps trying to upscale old footage . The other one that's really great is the AI motion smoothing, and this is great because sometimes the videos you get out of AI video generators can be a little bit jittery. And they also give you the ability to convert your video into super slow motion, so this is a great way to play with your AI video footage. As you can see from this before and after, it dramatically increases the quality and resolution of your AI video. The other set of important tools you'll want to use in your AI filmmaking process are LLMs, and these can be used for a whole range of different tasks throughout your creative AI journey. And I like to treat the AI chatbots as my companions, my sparring partner, and I often just sit here, I turn it on to chat, I'll go through my whole thought process, I'll lay it all out on the table, and then ask it for its feedback, its reflections, its organization and its input on how it can be improved.
So it's great for coming up with everything from ideas for your character names, to the name of your movie, to getting initial feedback on how you're progressing through the narrative of your story. And with ChatGPT you can upload images or stills from your video and ask for feedback on how they look, how they appear, what someone might interpret from these shots. And I also think it's a creative thing that we all often struggle with to get things done, and often I would ask ChatGPT to just do a little motivational speech .
My dear ChatGPT, I'm struggling to complete my latest creative project and I'm wondering if you could give me a little inspiration and motivation. It would mean a lot if you could encourage and help me on this journey. You have everything you need to succeed in your. Strength, passion and the will to take on any challenge. Remember that the journey to your goals is not just about the destination. And as you can see, there is a lot of support, because I think one of the most important things in the creative process is having inspiration, motivation, dedication, a goal and a thought about the bigger picture. That's why I often use ChatGPT to keep me on track. Another option for an AI chatbot is to try Claude. And I've also had a lot of fun using Claude, so I recommend you explore that as an alternative to ChatGPT. And it can also be interesting in some ways to ask both of them the same thing and see how they interpret the answer . Especially if you want to get feedback on your work.
So I think it's really interesting to get feedback from both tools on video titles, but also on movie titles, and see how they approach this from different perspectives. So for that, take a look at Claude . The use of AI now offers us a whole range of opportunities to be creative. And my invitation to you is: is this what interests you most? Because I believe that one of the most meaningful things we can do in our lives is create works of art. To create beautiful things. And my invitation to you is to follow the words of Steve Jobs. And what he's saying is, the only way to do great work is to love what you do. What I find interesting about this is that if we follow what we're most curious about, what interests us, what baffles us, what makes us curious, then we have the ability to create things that will be excellent . And if you're curious about AI filmmaking, check out the AI Filmmaker Academy, where we make films together. A special discount applies for this in the description below.
But most of all I want to say thank you so much for watching. Thank you for being here. And most of all, I wish you a nice day..
