我花了 $2.00，做出了一个声音逼真、喋喋不休的 AI 网红

我花了 $2.00，做出了一个声音逼真、喋喋不休的 AI 网红。 左边：真人创作者 右边：AI 网红 wan 3.0 prime > seedance 2.5 这套完整流程值得存下来： 用 GPT-Image 2.5 和这段提示词创建 AI 形象（按你想要的外貌调整）： “Ultra-realistic iPhone selfie portrait of a beautiful 22-year-old woman with soft European facial features, fair skin, subtle natural freckles, warm brown almond-shaped eyes, long lashes, a small straight nose, naturally full glossy pink lips, and softly defined cheekbones. Shoulder-length layered champagne-blonde hair with darker roots, voluminous curtain bangs, and slightly messy natural waves. Wearing a fitted dark charcoal long-sleeve top, small silver hoop earrings, and a beaded necklace with a round spiral pendant. Relaxed neutral expression, subtle closed-mouth smile, direct eye contact. Standing in a dim bedroom with beige walls, dark furniture, and hanging ivy in the corner. Soft frontal window light illuminating her face, dark underexposed background. Authentic casual iPhone photography, realistic skin texture, natural imperfections, no beauty filters, no AI look, vertical 3:4 composition.” 用这段提示词给你的 AI 形象生成角色设定图： “Character turnaround sheet of the person from the reference photo, six views: full body front, side and back + super close-up front, side and back. Strictly keep 100% of face, hair, body, clothing and all details from the reference. Clean gray background, consistent and natural lighting.” 下载一条热门 tiktok 或 reel，内容是创作者对着镜头说话 从视频中剪出一段 10s 片段 在 wan 3.0 prime 中，将视频和角色设定图作为参考，使用这段提示词： “Replace the girl in {video} with the girl in {character sheet}. Keep the same movements, camera quality, voice tone, background sounds, and hand gestures. Remove ALL on-screen text. NO subtitles, NO captions, NO text overlays at any point. Replace the original speech with this EXACT script, word for word. Pronounce every word clearly with natural pacing and accurate lip sync. Do not skip, change, rush, or cut off any words. Make sure the entire script is spoken. {script}” 想看更多流程就告诉我！