Deepfake technology is basically a technique in which algorithms are used to add some words to the speech of people which they did not originally speak. However, the videos sound so real that you wouldn’t imagine that this is a deep fake video.
What is Deepfake technology?
The best example of a deepfake video was recently seen in the form of Former US President Barack Obama’s speech. In this video, Barack Obama calls President Trump “total and complete dip****” which is obviously not what he said in reality. However, you might totally believe that Barack Obama said those things for President Trump if you don’t know what Deepfake is.
Also, Deepfake is a serious challenge for people who seemed to think that what they see in videos and image is real evidence. Talking about Deepfake technology in general, it can do a lot more things than just adding some content to the speech which was already given. It can even swap speech from someone on someone else’s face. So you might think that the speech was actually given by the person who is seen but the voice is from someone else.
GAN or generative adversarial network is the technique used to create deepfake content. GANs make use of two neural nets known as a generator and a discriminator. These two neural nets constantly compete with each other.
Talking about the competition between the generator and the discriminator, the generator generates a realistic image. On the other hand, the discriminator tries to determine if the image is fake or not.
If the generator creates a fake image and discriminator finds that it is fake, it will make an even better fake image. However, if the discriminator does not find a fake image from the generator then it will try harder to determine the fake image. This constant cycle continues until it is deemed that the image or video is no longer fake for humans.