معهد دبي القضائي

DJI Platform

Documentation

How it works

The steps

  1. Outline. From the brief, the system proposes a slide-by-slide outline the author can adjust.
  2. Slides. It writes each slide's content on the chosen template, detecting Arabic or English and laying it out accordingly.
  3. Narration. It reads each slide aloud in a selected voice, producing the audio.
  4. Video. It combines the slides and narration into a finished video, with optional intro/outro and background music.

A PowerPoint export of the deck is available at any point.

Templates and voices

Authors pick a template (colours, fonts, layout) and a voice. Voices are available in Arabic and English, and a custom voice can be set up from a short sample.

Saved as it goes

The work is saved slide by slide. If the author closes the page partway through, it picks up where it left off rather than starting over, and a problem on one slide doesn't stop the others.

Producing the video

Turning slides and narration into a video is the heavy step, so it runs in the background: the author submits the job and is notified when the finished video is ready, rather than waiting on the screen.