Create a localized version of Video 1 using the complete approved script in Text 1. Preserve the adult presenter's identity, age, skin texture, hair, wardrobe, body position, gestures, product geometry, hand-object contact, step order, camera, cuts, background, exposure, and total duration. Replace only the spoken language and the mouth movement needed to match it. Use Audio 1 only for the authorized speaker's vocal identity; keep a natural pace and pronunciation for the target locale.

The three product actions must still land on their original moments: unfold the handle, rotate the selector one click clockwise, and place the device on its base. Do not change or imply any feature beyond the approved script. Remove the source-language lower thirds cleanly and leave the three supplied graphic areas blank for exact typography to be composited from Images 1–3 after generation.

Mix the new dialogue at the same perceived distance as the source; preserve room tone and action sounds. Duck background music only when needed for intelligibility. Do not translate brand names, serial numbers, legal copy, or units unless Text 1 explicitly supplies the approved form. Avoid voice swapping, a second speaker, extra claims, paraphrased safety instructions, lip jitter, altered hands, baked-in generated text, new cuts, or watermark.