Summary
Julia McCoy walks through a four-step process for turning a single phone selfie into a finished, subtitled talking-head video using the Veed AI video platform and a cloned version of their own voice. The aim is to show business owners and creators how to produce short-form video without a studio, retakes, or a camera crew.
The tutorial covers the details that affect whether an AI avatar looks convincing. Use the back camera in natural daylight, face the camera straight on, keep the image in focus, and keep the mouth slightly open so the model renders teeth accurately. From there, the video shows how to write the prompt, clone a voice, choose a character, subtitles, duration, and audience, and turn off background music. It also recommends rewriting the AI-generated draft script before publishing and using the editor to add B-roll, screenshots, and sound effects.
McCoy contrasts this quick selfie-based version with a fully produced studio AI clone that took a voice-recording session and a production team and generated over a million impressions in a month. The simple version is presented as an on-ramp for short-form ads, quick customer answers, and a week of content from one sitting at a desk.
📺 Source: Julia McCoy · Published September 30, 2026
🏷️ Format: Tutorial Demo
