Text to video AI: from one sentence to a finished, narrated video

Text to video usually means a single short clip from a prompt. ShortsMachine treats your text as a brief for a whole video. "Three surprising facts about octopuses" becomes a titled script with a hook, several scenes and a payoff — each scene with its own AI-generated shot, narration and captions.

You can be as vague or as specific as you like. A topic works; a detailed outline with the points you want covered works better.

How to write text that becomes a good video

The script writer works from your idea, so the clearer the idea, the better the video. Three things make the biggest difference: the angle, the audience and the ending.

  • Angle: "Why the Roman Empire fell" is a topic; "Five mistakes that brought down Rome, told like a detective story" is an angle.
  • Audience: say who it is for — kids, beginners, investors, horror fans. Vocabulary and pacing change with it.
  • Ending: ask for a twist, a lesson, a question to the viewer or a call to follow. Good endings keep people watching to the last second.
  • Must-include points: list facts, names or steps that have to appear, in the order you want.

Example prompts that work

  • "A 60-second scary story about a lighthouse keeper who hears knocking from inside the lamp room. Slow build, twist ending."
  • "Explain compound interest to teenagers using a pizza analogy. Friendly, upbeat, 45 seconds."
  • "Five little-known facts about the Moon, each with a dramatic space visual. End by asking viewers which fact surprised them."
  • "A motivational story about a marathon runner who finished last, told in second person. Cinematic, sunrise lighting."

Style, voice, language and format

Before generating you choose an art style (cinematic, anime, watercolor, comic, 3D render and more depending on plan), one of up to 15 voices, the narration language (English, Spanish, French, German, Portuguese, Italian, Japanese, Korean, Chinese, Hindi or Arabic) and the format: 9:16, 16:9 or 1:1.

The style is applied to every scene through a shared visual style line, which is why the same character keeps the same clothes and colours from the first shot to the last.

Length: 15 seconds to 5 minutes

Pick 15s, 30s, 1, 1.5, 2, 3, 4 or 5 minutes. The script writer plans the number of scenes and the amount of narration per scene so the finished video lands close to that length. Short durations suit hooks and single facts; one to three minutes suits stories and explainers.

Frequently asked questions

Is text to video free?

Every plan starts with a 3-day free trial. After that, plans start at $3.99 a month with 200 credits, enough for about two and a half minutes of video on the cheapest models.

Can I write the script myself?

Today the script is generated from your prompt. If you include the exact lines you want in the prompt, the writer will usually keep them.

What languages does text to video support?

Narration and captions can be generated in 11 languages: English, Spanish, French, German, Portuguese, Italian, Japanese, Korean, Chinese, Hindi and Arabic.

Make your first video in a few minutes

Type an idea and ShortsMachine writes the script, generates every scene, adds a voiceover and captions, and stitches it into one video — up to 5 minutes long.

Start your 3-day free trial

Related guides

Text to Video AI: Turn a Sentence into a Narrated Video