Huggingface has been one of my new interest. They seem to have various AI based models that we can use for the generation of the video, image and the text. And it can offer you plenty of things that you can work around with for creating content. So it's worth exploring as much as possible.
The model I am making use of is listed here. So if you happen to make use of what I am trying out here. You should check out the link and see if that works out for you. It would give you an idea on how to work with the text to video here.
Image Credit: infotech report
Let's take a look at the prompt that generates the video out of the text that are providing. So here's my prompt for you to check out.
Woman driving the car in middle of the forest road at night. A creature standing in the middle of road blocking the road for the car. Woman applies breaks to the car, afraid and stops her movement looking at the car.
Let's see how the prompt turns out for me. Though I can't upload video here but I can atleast show you the screenshot so that you can get the idea of how those things work and how those things can work around with the same.
As you can see my experience has been extremely poor. Like say the text to video is basically scraping out of the video stock companies content. Like this one was copying out of the shutterstock. And it was not generating out of it's own mind or so.
So it's kind of obvious but the thing is that AI has been extremely hard to work around with the way things are going on.It is kind of open copying there. So in short we are at the point where things are not that good and we are just finding copy pasted content. So it's totally waste of resources on that part.