Text to Video Generator

Turn a written scene into a short video. Direct the subject, action, lighting and camera as text, then choose a supported clip length.

Results

Tripod-mounted phone framing a woman in a rust jacket vertically

Example

Your result will appear here.

See examples

Write a text-to-video prompt for one shot

Name the subject, add one action and finish with a camera instruction. A single timed action is easier to fit into a short clip than a list of scenes.

When to use it

Scene with visible geography

Use a locked wide shot when the street, doorway or horizon is important. Specify where in frame the action starts.

Close product or food detail

Use a push-in toward one feature. For a drink, describe condensation and a stable countertop before adding a pouring action.

Examples

Yellow tram turning through a rain-wet street at dusk — The tram, wet pavement and headlights give the prompt three concrete visual cues.

Cycle courier in a yellow jacket on a city street — A tracking shot follows the courier while leaving room ahead of the bicycle.

Example

A yellow tram sits at a bend in a wet street at dusk. Pan right to reveal the tracks and their reflections.

A pan over a still illustration shows the camera direction in the example prompt.

A steady frame lets the drinks and fruit suspended behind the subject carry the action. Pollo AI

Direct the same tram scene two ways
Shot	Prompt instruction	What to review
Locked wide shot	Keep the camera across the street as the tram rounds the corner.	The street and wet reflections remain in view.
Slow push-in	Move slowly toward the tram headlights as it finishes the turn.	The camera closes the distance smoothly without an extra cut.

How to use it

Write one sentence naming the subject, setting and visible action.
Add one camera direction and select a clip length that gives the action time to finish.
Generate and watch the first and last frame. If the model cut the action short, simplify the prompt before adding duration.

Settings & limits

Prompt: 1 to 2,500 characters
Duration: Depends on the video model
Output: One video clip

Questions about Text to Video Generator

What should a text-to-video prompt include?

Include one visible subject, one action, a setting and any lighting or camera instruction that changes the result. “A tram turns a wet corner in a wide shot” is easier to direct than a list of unrelated cinematic adjectives.

Why did the model change camera angles?

A sequence of different viewpoints or words such as montage can invite cuts. Ask for one continuous shot with a fixed camera or one specific move.

Can I write a prompt for a shot with dialogue?

Describe who is speaking and keep the spoken line short. Audio behavior depends on the selected model and controls. Use AI Voiceover and Add Audio to Video if you need a separate narration track you can review first.

When should I use Image to Video?

Use Image to Video when the exact starting composition matters, such as a product shot based on approved artwork. A text-only prompt leaves the initial appearance to the model.

What should I change when the clip feels rushed?

Remove secondary actions and describe the opening frame. For a five-second shot, let the tram finish one turn. Save a second reveal for another clip. If the first frame needs to match a particular composition, use Image to Video with a still.