Skip to the contentVideo tooling
Ami StudioOleksii Zahlukha

Video toolingPrisma — translate anything on your screen

Prisma — translate anything on your screen

A screen translator: select text in any window and the translation appears beside it. Plus text from images, PDFs and video, voice input, and translating a whole window at once.

Used for two different jobs: everyday reading and pulling text off video. It is filed under video tooling, where the second job lives.

What it is

A translator that lives on top of the whole system rather than inside its own window. That difference decides everything: the moment a person has to copy text into a separate application, they stop using it — by the second or third day. Here the translation comes to where the text already is: the browser, a document, somebody else’s program, a video player.

The text is taken three ways: by selection, by recognising it straight off the screen image, and through a frame laid over video — the last of which is what puts this work in the video tooling section. There is voice input and a “translate the whole window” mode, for when what needs translating is not a phrase but the interface of an unfamiliar program.

Specifications: Python, on-screen text recognition, local and free cloud models with switching between them — the translation costs nothing. It carries 121 of its own automated checks, run before every build.

Nearby work

Back to Video tooling