# How to make a video with a child in VEO 3

A trend is gaining popularity online where children "take on adult roles." We've put together a simple tutorial on how to recreate such a video using the VEO 3 neural network: from writing the scene in ChatGPT to generating the final video.

1. In the ChatGPT chat, write in detail in Russian what you want to see in the video — describe the scene, characters, their actions, and what they say. Dialogues should also be in Russian. Example:

“Create a scene: a small child is cooking food in a kindergarten, acting as a chef, and says: ‘The adults have fallen asleep, now I’m in charge in the kitchen. Cooking is so hard, no one helps.’ Let there be a huge smile on the face and a little flour on the nose. Around are dishes, a ladle in hand, and the child stands on a chair above a pot. Make the description detailed so that the child, the kitchen, and the background are all visible.”

1. Ask ChatGPT to turn this into a detailed prompt for the [VEO 3](<https://veo3.art/>) neural network, up to 2000 characters — so that the description is in English, but the dialogues remain in Russian.
1. Copy this ready-made prompt and paste it into VEO 3 to generate the video.
1. Enable a VPN with a server in the US before uploading the prompt.
