The owner of TNT and Rutube will set up a
Software

The owner of TNT and Rutube will set up a studio for content production using AI

Gazprom Media Holding (owns NTV, TNT and TV-3 channels, Central Partnership film company, Rutube hosting, etc.) will open an experimental studio for content production using neural networks. About this in conversation with journalists at the Eastern Economic Forum told the head of the digital laboratory of the D. Lab holding, Eduard Maas.

    Image source: steve_a_johnson / Pixabay

Image source: steve_a_johnson / Pixabay

According to available data, the studio will use about 20 artificial intelligence models, including open source projects (e.g. the Stable Diffusion image generator and the LLaMA language model) and Gazprom-Media’s own developments. The studio will produce animations using generative algorithms, as well as create virtual presenters, process archive content and work on special advertising projects. The amount of investment allocated to the development of the project was not disclosed.

Mr. Mass believes that content creation technologies using generative neural networks will be useful for production studios, advertising agencies and bloggers, most notably using the video hosting Rutube and the social network Yappy. In addition, the holding will offer its technologies to external customers, including representatives of the banking sector and state-owned companies. The company is currently in discussions with one of the largest banks in marketing and a large publishing house where generative algorithms can be used to illustrate books. The holding itself can use these technologies to create advertising on social networks, generate animations from audio, create virtual presenters, etc.

Maas made it clear that the proposed technologies would not be used to produce “a hit series or a top show.” At the same time, artificial intelligence can significantly reduce the costs of individual production processes. Such technologies can be used, for example, to render backgrounds, characters and video details. Neural networks can also be useful for completing vertical videos, making them horizontal, synchronizing works with realistic synthesized voices, etc.

About the author

Robbie Elmers

Robbie Elmers is a staff writer for Tech News Space, covering software, applications and services.

Add Comment

Click here to post a comment