I’ve been playing around for a few days now with how to use AI applications to create.
As you know, in my last post AI Future I wrote about how the world could look like in 3-5 years. Well, out of the 900 AI applications, I was the first to start looking at how to optimize content, whether in written form (blog) or in video form (educational videos, podcasts, etc..).
So far I have come to this, after some testing. Of course it needs to be tweaked, but the process seems to be “simple” for a normally capable person with a computer, who doesn’t have to be a geek. So I don’t consider myself a geek.
Text in written form:
I have text in written form and would like to put it into a video and create a video.
The best way is to use openai’s chatGPT to revise your text and throw out unnecessary words. Maybe you can let the text transcribed, so that it is more condensed and with only brief points.
Then use Deepl and have the text translated into whatever language you want. Of course, it is good to know the languages, especially if it is a specific term or a rather technical context, where Deepl makes mistakes.
With the help of synthesia AI you can generate a video with your clearly given text.
Voila, you have a cool video with an AI avatar, in about 100 different languages.
If you pay an extra $1000/year, you can create your avatar that looks like you, talks like you, and then you can have the AI work to create a video of your avatar with the text you provided.
Voice audio recording:
If you have an audio recording, such as a podcast, share, etc… the easiest way to do this is to have the AI transcribe the recording to text. Descript is great and can do other useful things, but I’ll cover that another time.
Then you will follow the same steps as in the previous paragraph for the text in written form.