Digital Media and Artificial Intelligence (AI): Applications for sounds and pictures

Back in the 1950s, the pioneers of artificial intelligence (AI) assumed that learning and artificial intelligence (AI) could be simulated by a machine. Particularly since the 2000s, the numerous projects, research and application developments testify, on the one hand, to the growth of this IT sector, and, on the other, to the major human and financial investments made by the world's leading players in the development of projects and products incorporating AI. In the United States: Google, Apple, Facebook, Amazon, Microsoft – and in China: Baidu, Alibaba, Tencent...

Another characteristic of AI is the wide range of knowledge and technologies involved: cognitive sciences, learning modes and machine learning, automatic speech processing, signal and image analysis and processing, computer vision, robotics...

The aim of this series of two articles is to provide an overview of the quantity and diversity of AI applications in digital media, which have been multiplying since the mid-2010s.

This first article is divided into three parts:

• a review of the principles, components and techniques of AI, as well as its uses;

• a sample of AI applications in the field of images (photo, film, video);

• a sample of AI applications in the field of sound (speech, 3D audio, music).

The second article [TE 5 898] :

• presents these and other AI applications in the broadcast and media industry;

• focuses on two case studies:

  • journalism and AI,

  • deepfakes.