Getting Started
A first project, from the Home page to an exported MP4. For an overview of what Scenema makes and how the pieces fit together, start with the docs home.
1. Sign in
Open scenema.ai and sign in with Google or Discord. Your workspace is created on first sign in, on the Free plan with 200 credits to try the whole studio: a video of about 30 seconds at Standard quality (768p), with a watermark. See Billing for the paid plans.
2. Choose what to make
The composer on the Home page has two modes:
- Explainer video. A narrated, animated video for explainers, ads, tutorials and training modules.
- Text to speech. Voiceover audio from a script, in more than 75 languages.
3. Describe it, or bring what you have
Type a short prompt. A sentence or two is enough, because the Director expands it into a full script. For example:
A two minute explainer on how vaccines train the immune system, for high school students.
You can also paste or upload your own script, attach documents, slides, images, spreadsheets, audio or a YouTube video with Add files, or mention a file from your library with @. Attached files are read once and kept in your library.
4. Set the length and the shape
Before you start, you can pick:
- Duration. Auto, or a target length from 15 seconds up. On the Free plan a video runs about 30 seconds.
- Shape. Horizontal (16:9) or vertical (9:16). Leave it for later and the Director asks you for it before the shots are planned.
Then start the project. The studio opens with the Director in the chat.
5. Answer the Director’s questions
The Director asks the main choices on cards right above the chat input, one at a time, as the video needs each one:
- Shape. Horizontal or vertical, when you did not pick it on the Home page.
- Style. A few styles that fit your idea, each with a sample keyframe. Browse all styles opens the full catalog, with your own styles. Creator and Pro can create a style from images, YouTube links of videos in the look they want, or a description.
- Narrator. A few voices to play and pick from. Show more voices opens every voice and language.
- Script. The script the Director wrote, to read in full. Record it records the narration; Ask for changes lets you say what should change.
Your pick is applied as soon as you answer, and the Director goes on from it. You can always write in the chat instead: the card closes and the Director reads your message. Answered cards stay in the chat history, greyed, with what you picked.
6. Shape the narration
The Director writes the script from your prompt and files, or uses your script as it is, and shows it on the script card. Each recording is kept as a take, so you can go back to an earlier one.
If you paste a storyboard with labelled voiceover lines (Voiceover:, VO: or Narrator: next to scene headings and visual directions), only the voiceover lines are narrated.
Your script becomes one recording, and the shots follow it: your scenes keep their order and their words, timed to the narration.
The narration is real speech in a supported language. Scenema does not make sound effects, ambience, non verbal sounds such as groans or breathing, or invented languages. Describe sounds in the shots instead. When a chapter’s narration cannot be made because of such a request, the project is locked: nothing more can be generated or edited in it, and you can try again in a new project from the home page.
When the narration is ready, an approval card opens in the chat with the recording to play. The chat input waits until you answer it. Listen to all of it, then approve it, or choose Fix something and say what should change. Until you approve, you can change the script, the voice and the recording as often as you like. The shots are planned only after you approve.
After the approval, one more card asks how to go on. Auto lets the Director run the rest of the production and email you when the video is ready or needs a decision. Guided keeps you in charge of every step, and is recommended for videos longer than 5 minutes.
Approving makes the script final. Its words cannot change and its recording cannot be replaced afterwards. You can still have it narrated in other languages: ask the Director, for example “narrate it in Spanish”. A chapter can have up to 10 translations. The script step shows one language at a time: pick the approved script or a translation above it. In a translation you can fix a word or a name before it is recorded. A change that makes it much longer or shorter is not saved. Each language is a take of its own, so you can switch between them in the takes. To use different words, start a new project.
A translation is fitted to the chapter’s shots: Scenema speeds it up or slows it down a little to their lengths. A narration that would play more than 15% faster or slower would not sound natural, so the chapter keeps its earlier narration and the studio says why.
Your own recording needs no approval: its transcript becomes the script. You can replace it until the shots are planned. After that it stays, and no new narration is generated for the chapter. A music track can be replaced until the chapter’s first keyframe is drawn.
7. Review the plan and the keyframes
The Director plans each chapter: the recurring characters and places, a stage for every place, and a keyframe for every shot, timed to the narration. Keyframes are drawn in your style, so the same people and places look the same throughout. To change what a shot shows, ask the Director in the chat; only the shots the change touches are planned and drawn again. Redraw any image until it is right.
8. Animate
Choose Animate and a quality: Draft (480p), a quick preview; Standard (768p), finer text and face rendering; or High Definition (1080p), crisp text rendering and fine details for commercial and broadcast use. You can pick up to your plan’s limit. Standard is where every project starts. Animate one shot, a few or the whole chapter. Animation is optional: a shot without a clip plays its keyframe with a slow pan and zoom.
9. Captions
Captions show the narration’s words over the video, timed to the narration and in its language. Turn them on in the studio or ask the Director, and restyle or hide them at any time. Subtitles are always shown as captions and never drawn into the images. Titles, signs and labels that are part of a scene are drawn as usual.
10. Export
Choose Export. The video renders as an MP4 at the project’s highest shot resolution, with captions if they are on. Pick which chapters to include; Creator and Pro can also download every shot’s clip as a zip. Renders usually take 5 to 15 minutes, and Scenema emails you when the video is ready. Download it from the timeline or the Exports page.
In a guided project, once every shot is animated, the Director asks on an export card: which chapters, and with audio or silent. On the Free plan the video carries the watermark; Without watermark opens the plans. When the render is done, the finished video appears in the chat with its download buttons below it.
A video exports one render at a time. Exporting again while a render runs, from another tab for example, follows the render already on its way. A chapter’s narration also records one take at a time.
Next steps
- Pricing FAQ: credits, plans and what each one includes.
- Workspaces: make videos with a team.
- MCP Integration: make videos from Claude, ChatGPT or Cursor.