Fifteen modules on Digen AI — every menu explained, then image-to-video, text-to-video, talking avatars, motion transfer, image editing, Agent skills, Real Motion 3.0, guest models like Sora and Veo, prompt craft and credits. Written to teach from, not to skim.
Now rendering
Image → Video · 4s · 720P
The whole workflow
The course
Teach them in order for a full unit, or pull a single module for a one-off workshop. Every module is a numbered workflow you can run live.
Run these three together, on one screen, before anyone generates.
A browser studio where a photo, a script or a reference clip becomes finished video — no timeline to learn.
4 steps
Do this once as a class, together, before anyone starts generating. Find the balance, note the number.
4 steps
Every menu in the top bar and every control under the prompt box, explained left to right.
6 steps
The four tools that make moving pictures.
A still photo becomes a short clip with realistic motion. The fastest first result in the whole course.
6 steps
No source image — a written prompt or short script becomes a scene. This is where prompt craft starts to matter.
5 steps
A still photo becomes a talking presenter, synced to a typed script or an uploaded audio track.
6 steps
Take the movement from a reference video and retarget it onto a single photo. No mocap, no green screen.
4 steps
Build the still first — it is the cheapest place to iterate.
Borrowed pipelines, the flagship engine, and the guest models.
Packaged, creator-published workflows plus a community feed — the two fastest ways to get unstuck.
4 steps
Digen's own flagship engine: motion, dialogue, ambience and music generated together in one pass.
5 steps
Sora 2, Veo 3.1, Runway, Nano Banana, Flux and more, all inside the same model selector.
3 steps
The part that transfers to every other AI tool.
Credits as a production budget.
A five-session progression plus a capstone.
Learning path
Level 00
Open the account, read the top bar, ship one clip in the first hour.
6 topics
Level 01
The four core generators, and how to keep a character looking like itself.
8 topics
Level 02
Shots become sequences. Continuity, lighting and lens language.
8 topics
Level 03
Full productions with a brief, a budget and a delivery deadline.
7 topics
Inside the academy
Movie Maker
Idea, story, characters, scenes, images, motion, voice, music, final cut — nine stages with the continuity rules that hold.
Open Skills
Sixteen skills — action, horror, documentary, ads, character consistency, storyboards — each with a working prompt and its steps.
Prompt Academy
The eight parts of a shot description, an interactive builder, and a tool that breaks any pasted prompt apart.
Camera Academy
Twelve shot sizes, ten camera moves, and the exact prompt wording that produces each one.
Style library
Cinematic, documentary, anime, editorial, Indian cinema and more — as copyable style blocks.
Prompt library
By type, style and industry — real estate, restaurant, fashion, education, healthcare, automotive.
Project-based learning
Featured prompts
Advertisement · Technology
A matte-black wireless earbud case rests on brushed steel, slow orbit right, single hard key light with soft rim, dust motes in the beam, shallow depth of field, premium tech commercial, cool neutral grade
Video · Restaurant
Steam rises from a clay bowl of biryani, camera pushes in slowly over the rim, warm tungsten kitchen light, saffron and gold tones, hands lifting the lid at frame edge, appetising food commercial
Video · Real estate
Wide interior of a sunlit apartment living room, camera glides forward through the doorway, sheer curtains breathing, late afternoon light across pale oak floors, clean architectural grade, no people
Courses
Reference
The five-part formula, weak prompts rewritten as shot descriptions, and patterns to copy in class.
Reference
Every generation mode side by side, plus Real Motion 3.0 and the guest engines.
Reference
Every term a student will hear — Act-Motion, Agent skill, Space, credits — in plain English.
Questions every class asks
Check Space first — failed jobs sometimes still appear there. If the clip is genuinely missing, re-run it once; if it fails again, change the model rather than repeating the same prompt, and report it from the account menu.
Because words alone cannot hold a face. Save one portrait as the reference and upload it as the source image for every shot, repeating the same physical description verbatim.
No. Most modes output silent video unless you supply a voice track or use a model that generates native audio. Make the voice separately and lay it over the clip.
It depends on the mode. Lip-sync avatar work runs longest; the cinematic video models produce a few seconds per generation. Build long pieces by cutting several short clips together.
Assignment examples
Three shapes of brief students will hand in — each one traces back to a single module.
Talking-head explainer
Avatar + lip sync · Module 06
Product b-roll loop
Image → Video · Module 04
Cinematic street shot
Real Motion 3.0 · Module 10
These are reference frames generated for the guide, not student work. Send real submissions and they can replace them.