Bedtime stories
Write down the story you tell your kids and watch it play back with a narrator, characters who talk, and soft music under the quiet parts.
Write the way you normally write, a paragraph, a story or a page of notes, and get back a video with pictures, voices, music and sound already in place.
Try it with your own words: your first video is on us, one minute of storyboard frames animated and voiced.

Most text to video AI tools read your words as a single prompt and answer with a single moment of motion. filmaris reads them the way a production reads a page. A loose paragraph, a finished short story or a list of half-formed notes is first shaped into a script. Every scene is then broken into shots, and each shot is filmed by a video model. Characters speak with ElevenLabs voices, ElevenLabs Music writes the score, and sound effects fill the space around them: rain on a roof, a door, footsteps. The whole mix lands on one timeline, so what you open at the end is a video, not a box of parts.

In text to video generation, no model is equally good at everything, and a story asks for many kinds of shots. So filmaris keeps a whole rack of engines and decides shot by shot which one renders that moment best. Among the current ones are Veo 3.1 from Google, Kling 3.0, Hailuo 2.3, Wan 2.7 and Seedance 2.5, part of more than fifty models that also cover pictures, sound and language. You write once. filmaris plans the framing and camera for each moment of your text, hands it to the engine that suits it, and gathers the results back into one continuous video rather than a row of unrelated clips.

filmaris opens like any web page, so a laptop at a café table is enough of a studio. Paste a story you wrote on the train, or type it straight in, and see what it looks like before you spend anything. With your e-mail, the script filmaris writes from your text costs nothing, and neither do the first images of its characters, places and props. That alone shows whether your words carry a video. Your first finished video is free too, made from those pictures in motion, and when you want every shot rendered as true AI footage, a paid plan takes it the rest of the way.

One line, a paragraph, a full story or bullet-point notes all work. Say what happens and who is there, and filmaris builds the script around whatever you give it.
Each moment is planned as a shot and rendered by the engine that suits it, characters get their voices, music and effects are added. Check every stage, or let it run alone.
The video opens on a timeline in your browser. Shorten a shot, swap two scenes, turn the music down under a line, then save the finished file.
Write down the story you tell your kids and watch it play back with a narrator, characters who talk, and soft music under the quiet parts.
Turn the backstory of a game world or a tabletop campaign into a short video your players can watch before the first session starts.
Write each episode as a few paragraphs and get a voiced, scored video in which the same faces come back week after week, ready for your channel.
Turn a one-page treatment into a moving pitch, so a producer or a client sees the tone, hears the sound and feels the pacing instead of reading about it.
Paste a text and filmaris writes it into a script and paints the first pictures of who and what is in it, at no cost. Your first video, one minute long, is free as well: those images animated with camera motion and voices, yours to keep as a file on your own machine. Footage rendered shot by shot by the video engines comes with a paid plan.
Turn my text into a video →Artificial intelligence text to video means software that reads written words and produces moving pictures from them. A typical AI video generator turns text to video one short clip at a time. filmaris treats your text as a story: it writes a script, films each shot with a fitting engine and gives it voices, a score and effects.
Yes. Paste or type your text and you get its script and its first pictures at no charge. Your first video is free too: one minute, built from those pictures with motion and voices. Every shot as generated AI footage is a paid-plan feature.
It depends on the shot, which is why filmaris does not commit to one engine. It draws on Seedance 2.5, Kling 3.0, Wan 2.7, Veo 3.1, Hailuo 2.3 and others, and gives each moment of your text to the engine best suited to it.
Yes. Veo 3.1, Google's video model, is one of the engines filmaris can send a shot to. It sits next to Kling, Seedance, Hailuo and Wan, and the choice between them is made shot by shot for your video.
Almost any text that describes something happening: a paragraph, a short story, a diary entry, a scene idea in bullet points. To convert text to video, paste it in and a script is written from it. Finished screenplays and whole books can go in as well.
Yes. Characters and narrators speak with ElevenLabs voices, ElevenLabs Music composes the score, and effects such as wind or traffic are generated per scene. Everything is mixed on the timeline, where you can change levels or add a music track or footage of your own.