Mode 01 • System
Speech to video
We turn spoken sentences into evolving visuals in real-time, collected using a microphone.
Discover the setup
PORTAL
An art installation that enables the real-time generation of a continuous video from written or spoken words.
The Technology
Developed in house, our technology combines real-time AI generation, signal optimization and artistic expertise to create interactive visual experiences that create emotions.
System
Mode 02 • System
Text to video
We generate a continuous video from queued text prompts written by the audience.
Discover the setup
Explore
Mode 01 • Speech to video
Speech to video
Our speech to video setup lets people command an AI algorithm by speaking inside a microphone.
Once a sentence is spoken, the video immediately transitions into a visual interpretation of it, then continues in constant motion until a new sentence is voiced.
- The experience can be presented on any type of screen or projection
- An automated moderation allows for content safety
- The installation only requires an internet connection
Mode 02 • Text to video
Text to video
Our text to video system is accessible through a web app, with no download required. Users submit text descriptions that directly steer the AI in real time.
Prompts are queued, enabling large scale participation. Each participant can track their position in the queue and see which prompt is currently shaping the video. As a prompt reaches the top, the continuous video stream gradually evolves in response, making each contribution visible and shared.
- The experience can be presented on any type of screen or projection
- The automated + manual moderation allows for 100% safety of the content
- The installation only requires an internet connection
Moderation
The installation comes with three levels of moderation
Natural moderation: the algorithms used for the visual creation are trained on a curated dataset, allowing for the exclusion of the content. We take advantage of the natural bias of the algorithms to ensure content safety without restricting the creative possibilities.
Style
We can adapt the style of the generation with your artistic direction. It can match some color elements but also some patterns. All we need is a few inspiration images to adapt the style.
Want to design your own experience?
We built it, we can adapt it.
About Us
We are Obvious, a French trio of artists and researchers working with artificial intelligence to create art. Inspired by the Renaissance workshops, we operate at the crossroads of academic research and art.
Our work consists in building artificial intelligence algorithms in the creative field, and producing artworks using those tools. Our research laboratory hosted in Sorbonne University and funded by the French National Research Agency (ANR) is leading research in the fields of image, video and sound generation. We are behind the first artwork created using artificial intelligence to go through a major auction house (Christie's, 2018), and our work has been exhibited in some of the world's most iconic museums.
