Company uses AI to create videos with your face

Imagine that you own a multinational and you need to speak with several employees in several languages, or you need to give a speech, but you don’t have a good camera and microphone available at the moment. To solve these problems, the Israeli company D-ID created the website Creative Reality Studio. In it, you can create videos of people talking and gesturing with just the image of their face.

Read more: Artificial Intelligence guarantees many benefits and offers a lot of utility

Creative Reality Studio Platform

The Creative Reality Studio tool has been in high demand, especially by multinationals. The need to have clear and assertive communication with employees around the world, without losing humanized communication, is the main factor for the success of artificial intelligence.

This is because, although the default language is American English, there are a total of 119 languages ​​available. Brazilian Portuguese is available in both female and male voices. Find out more information about the program:

How it works?

Basically, the platform has pre-made designs of bodies that will serve as a “template” for the face you want to insert there. Then, you load the image of the face of interest, and the program will place that face in one of the body models it already has in its database. This face in the photo will also move, either to smile or to simulate the movement of facial muscles during speech.

And how is the voice created?

The platform provides different female and male voices. However, for Brazilian Portuguese, unfortunately there is only one male and one female voice option. However, it is possible to upload an audio with the voice of the person who will participate in the video, to add more realism to the video.

How to avoid “deepfake” in this tool?

Many users may be malicious when handling the tool. Politicians, especially during elections, can be victims of slander and defamation. For this reason, the platform has some rules. In several languages, swearing and racist expressions, for example, were banned. In addition, the site expressly prohibits the use of photos of various politicians and famous people, such as Bill Clinton, for example.

They have a monitoring policy that bans users from the platform who may violate the rules of use for business purposes or, when personal, are offensive.

How to tell if the video is manipulated?

There are some clear signs. The person’s mouth does not always move correctly. Some movements are often artificial, choreographed, as the tool creates stereotyped facial movements to bring the image to life. In general, it’s pretty straightforward to figure out when it’s a legitimate video and when it’s manipulated.

Source link

About Admin

Check Also

ChatGPT: new version announced by OpenAI makes less mistakes – 01/30/2023 – Tec

New version of ChatGPT text generator robot should make fewer factual errors and better interpret …

Leave a Reply

Your email address will not be published. Required fields are marked *