About us

Everasa is a research lab that uses AI to understand films and television, and puts that understanding to work on accessibility.

Who we are

We are researchers in computer vision and machine learning with more than a decade of work on narrative media. Understanding a story on screen takes more than recognising what is in each frame. It means knowing who each character is, what they are doing and why, how they feel about one another, and how a plot builds across scenes and episodes.

That is where our research has focused: teaching machines to follow a story the way people do, from its characters and their relationships to the events that carry it forward.

Why accessibility

Audio description is the narration that tells people who are blind or have low vision what is happening on screen. Doing it well is a test of exactly this kind of understanding: a description has to follow the story, name the right characters and fit into the pauses between dialogue. Writing it by hand is slow and costly, so most films and series are still released without it.

So we built a service that writes it. Studios, broadcasters and streaming platforms upload a film or episode through their Everasa account and receive a timed audio description script, ready for their describers to refine and voice.

Our mission

To make every film and series accessible to people who are blind or have low vision, by building AI that truly understands stories.

To get in touch, visit our Contact us page.