Upload a few photos and a recording, and let their voice and face finish the words that were left unsaid.
Submit 1–5 photos — AI extracts the clearest front-facing likeness, and you keep the one you choose.
Upload a recording or video — we denoise it and pick the clearest segment as the voice preservation template.
Write the lines, pick a backdrop, and they'll say them to you in a familiar voice and face. Each clip is up to 8 seconds.