Do I need to write a prompt?

No. You talk to the chat the way you would brief a person: what the video is about, who it is for, how long. The sources you connected do the heavy lifting, and the script comes back for you to read before anything is rendered.

Can I use my own script?

Yes. Paste it into the chat or the script node and the voice-over and shots are generated from it as written.

Either an AI voice you pick from the list, or a voice cloned from a reference video you provide. Cloned voices are stored per user, so the ones you make stay yours.

Every plan includes monthly credits and generation spends them per action. Plans start at $13 a month; see the pricing section on the home page for what each one includes.

Make a video

An AI video generator that starts from your research, not from a blank prompt.

Drop the TikToks, YouTube videos, pages and PDFs you are learning from, ask for the video you want, and Infica builds it as a chain: sources → script → voice-over → shots → subtitled cut. Every step is a node on the canvas, so changing the script does not mean regenerating the whole video.

Plans from $13/month · credits included

What you get back

A vertical or horizontal video with a spoken voice-over, generated shots that follow the script scene by scene, and subtitles burned into the export. The script, the voice track and every shot stay on the board as separate nodes, each one editable and regenerable on its own.

Script written from your own sources, in the language you set for the chat
Voice-over from an AI voice you pick, or a voice cloned from a reference video
Generated shots per scene, with a plan that avoids empty or black frames
Subtitles styled and burned in — no separate captioning tool
A timeline editor to trim, reorder or swap any clip before export

How it works

1Drop your sources onto the canvas: links, video files, web pages, PDFs, images or audio. Each becomes a source node that is read once and reused by every chat connected to it.
2Ask the chat for the video: the topic, the length, the format and who it is for. If you connect a reference video, its pacing and structure inform the plan.
3Review the script node. Edit any line; only the steps downstream of what you changed will re-run when you ask.
4Pick the voice and the visual style, then generate. Shots, voice and subtitles come together in a final video node.
5Open the timeline editor if you want to change a cut by hand, then export.

Why a canvas instead of a prompt box

Single-shot video generators start at a text prompt and end at one clip. That is fine for a single shot; it is not how a video that says something gets made. The script needs sources, the voice needs a script, the shots need both, and you will want to change one of them after seeing the result.

On the canvas those dependencies are drawn as edges. Rewrite the script and regenerate only the voice. Swap the voice and keep the shots. Replace one shot without touching the rest. Nothing downstream re-runs unless you ask.

Formats and lengths

Vertical 9:16 for TikTok, Reels and Shorts, or 16:9 for YouTube. When you clone a reference, the output follows the format of that reference unless you pick one yourself. Short-form is the sweet spot; longer scripts are produced as a sequence of scenes and voiced in parallel chunks.

Questions

Do I need to write a prompt?: No. You talk to the chat the way you would brief a person: what the video is about, who it is for, how long. The sources you connected do the heavy lifting, and the script comes back for you to read before anything is rendered.
Can I use my own script?: Yes. Paste it into the chat or the script node and the voice-over and shots are generated from it as written.
Whose voice is used?: Either an AI voice you pick from the list, or a voice cloned from a reference video you provide. Cloned voices are stored per user, so the ones you make stay yours.
What does it cost?: Every plan includes monthly credits and generation spends them per action. Plans start at $13 a month; see the pricing section on the home page for what each one includes.