Sponsored by
How to convert text into cinematic AI video scenes in under 10 minutes
One of our favorite features of the Adobe Firefly creative suite is its AI video maker—a powerful tool that can help bring creative ideas to life quickly, and with surprisingly realistic results.
All you need is an Adobe account and a strong, cinematic text prompt and you can create a video that’s safe for use across your YouTube channel, marketing campaign or any other creative use.
Getting started on Adobe Firefly
You’ll need an Adobe account to access the text to video AI models, but even a free account will give you access to Firefly’s basic video generator.
As we’re focusing on commercially safe videos in this article, the Firefly Video AI model is all that you’ll need—while other models may give you more artistic control, Firefly Video is the only one that Adobe has labelled as safe for commercial use, so it’s the one we would recommend.
If your video is quite complex, or you have lots of ideas to generate, you might want to consider upgrading to a Pro plan (which includes a free trial). This will give you more credits to create a wider range of videos and refine your ideas.
Once you’ve logged into your account, head to Adobe Firefly’s video page, and select ‘Generate Video’.
Set your video parameters
When you open up the video generator, you’ll see a large prompt box in the center and some general settings to the left-hand side.
As tempting as it might be to dive straight into your prompt, spend a few minutes making sure you have the right settings first. This includes choosing the right AI model, selecting the resolution for your video, and choosing the aspect ratio.
The exact options available to you will change depending on which AI model you choose. With Firefly Video you will be able to create widescreen, square or vertical videos.
Optional advanced controls
The settings above should be enough if you’re creating a straightforward video. However, Firefly also offers several more advanced options that give you greater control over the final result.
This includes:
- Frames: upload a photo to act as the first or last frame of the video. This is particularly helpful if you have a clear brand identity that you need to bring to life, or existing creative assets that can act as the basis for your video.
- Composition: share an existing video (under 20 seconds long) and Firefly will analyze the composition of the video—components such as the layout, physical structure, and spatial arrangement—and replicate these in the video that it produces.
- Camera motion: similar to the composition option above, you can upload a short existing clip and Firefly will replicate the camera movement. This means that you can capture elements of existing clips that you love and replicate them in your new video.
- Style: there are a range of pre-set art styles, from cinematic to stop motion, and selecting one here helps to tell the AI video generator exactly what you’re looking for.
Write a great prompt (or get Firefly to do it for you)
Now for the fun part. The success of your video lives or dies on how good your prompt is, so the more specific and detailed you can be, the better the result. A good prompt should vividly describe what you want to see in your video, including the subject, what it’s doing and how it moves, and the background or setting.
To ensure a cinematic end result, you will also want to give instructions about details such as the lighting and camera movement. If you have experience with filmmaking, you may feel comfortable fleshing out those details yourself. If not, don’t worry. You can use the ‘enhance prompt’ function to turn a simple one- or two-line prompt into something rich in cinematic detail.
Here’s how the “Enhance Prompt” feature transformed one of our basic prompts.
Original: A close-up of a barista's hands as they skilfully create a foam heart on the top of a freshly brewed latte.
Enhanced version: A close-up of a barista's hands as they skilfully create a foam heart on the top of a freshly brewed latte. The barista, wearing a clean white apron, uses a metal pitcher to pour steamed milk, which forms a delicate heart shape on the surface. The setting is a modern café with sleek countertops and a backdrop of coffee-making tools. The lighting is warm and inviting, highlighting the rich brown of the coffee and the creamy white of the milk. The video is shot from a steady overhead angle, focusing on the precision and artistry involved. The overall style is clean and professional, emphasizing the beauty and craftsmanship of latte art.
You can see the video produced as a result below.
Generate and adjust your video
Happy with your prompt? Then it’s time to hit generate and wait for the AI video generator to do its job. In most cases this should only take a couple of minutes maximum; while we were testing the process for this article, it never took more than two minutes for the video to appear.
At this point, you may need to get comfortable with a little trial and error. You probably have a clear idea of the end result that you’d like to see, and there are usually some adjustments involved in getting there.
This could involve tweaking the prompt if something is missing or being misinterpreted, or using the “Prompt to Modify Clip” function to describe the specific changes you’d like to make.
Editing your clip with prompts uses additional credits, so you may need a paid plan for this step.
Export or edit further
For a straightforward video, you should be able to get what you need from just a couple of rounds of prompting and editing. In this case, the final step is simply to download the video and use it across your channels and campaigns.
If you have a more complex idea, then you may want to use Firefly’s Timeline Video Editor to edit it together with additional clips, or make other simple changes such as adding text and captions.
Alternatively, you can export the file to a more heavyweight video editing tool such as Premiere Pro to carry out more advanced editing.