Text to Video: How to Generate a Video From a Prompt
To make a video from text, describe the scene, subject, motion, and style in a prompt, choose the length and aspect ratio, and generate. Text-to-video is what you use when you have no footage at all — for B-roll, intros, concept clips, and faceless content. Keep prompts concrete and clips short for the most usable results.
What text-to-video is
Text-to-video generates a moving clip from a written description alone — no source image or footage. It is the counterpart to image-to-video: reach for it when you have nothing to start from and need a scene created from scratch.
→ Have a photo? Use image-to-videoWhen to use it
Use it for B-roll and cutaways, intros and transitions, concept and mood shots, and faceless content where no real footage exists. When you already have a product photo or portrait, image-to-video or lip-sync usually gives more control.
Writing a video prompt
Describe scene + subject + motion + style, e.g. "coffee beans falling into a white cup, slow motion, soft daylight, shallow depth of field". Name the camera move (push-in, pan, orbit) and keep it to one clear action per clip.
Length & aspect ratio
Keep clips short — a few seconds of clean motion is more usable than a long, unstable one. Use 9:16 for TikTok, Reels, and Shorts; 1:1 for feed; 16:9 for landscape. Generate at the ratio you will publish.
Faceless & B-roll content
Text-to-video is a staple for faceless channels: generate several short scenes, stitch them together, and add music or a voiceover. Combine with lip-sync for a presenter and image-to-video for product shots to assemble a full piece from parts.
→ Add a talking presenterTry it yourself
Generate a short video from a text prompt with AI: describe the scene, motion, and style, pick length and aspect ratio, and get a clip — ideal for B-roll, intros, and faceless content.
Try text to videoFAQ
How is it different from image-to-video?
Text-to-video creates a clip from words with no source image; image-to-video animates a still photo you upload.
How long can a clip be?
Short clips of a few seconds work best; generate several and stitch them for a longer piece.
What makes a good video prompt?
Scene + subject + one clear motion + style, with a named camera move; keep each clip to a single action.