Text to AI Video

Descript’s Text to AI Video tool lets you generate a complete video — including voiceover with a stock AI Speaker, visuals, a script and even the option to add an avatar — from a single text prompt. This tool lives inside the Descript Home tab and allows you to create narrated, illustrated videos without needing to write a script, record audio, or design visuals yourself.

Use this tool to turn text into video, create a video from an idea, or explore AI-powered video creation directly within Descript.

TexttoAIVIdeo.gif

Before getting started
  • Start with an idea, not a script: This feature works best when you give it a general idea—not a full script. If you have an existing script and just want to generate images, see Generate AI Images using Underlord
  • Processing time: The video generation process may take a few minutes to complete, depending on the length of your script and the complexity of your visuals. Please allow time for the AI to work its magic.
  • AI Actions and Text-to-speech usage: Using the AI Video Maker only counts as one basic AI suite action and deducts from your available TTS allowance based on the length of the generated script. Because the script is generated during the generations process, TTS usage can't be estimated in advance.
  • Images only: Currently, the AI Video Maker can only generate still images.

How to use Text to AI Video

  1. Open Descript and/or navigate to the Home tab. In the Popular Features section, click on Text to AI Video.
  2. In the prompt modal, describe your idea for a video.  ("Create a video about the origins of the Welsh Corgi," "Tell a short story about a time-traveling cat," Explain how honeybees communicate through dancing," etc.)
  3. Choose a visual style:
    • Low Polygon 3D – Modern, blocky, and video game-like.
    • Plasticine – Claymation-style, colorful, and textured. 
    • Watercolor and Ink – Soft, hand-painted, and storybook-like.
    • Whiteboard Doodles – Simple, sketchy, and playful.
  4. Then, select your preferred aspect ratio (Landscape 16:9, Portrait 9:16, or Square 1:1).
  5. If you'd like to include an avatar, toggle the Use avatars  option on.
  6. Click Generate AI Video. Descript will create a new composition that includes a written and narrated script in your default AI Speaker with stylized, AI-generated visuals based on the selected visual style. prompt

Once your video is generated, you can edit the script, replace visuals, and enhance the composition using Descript’s full editing tools.