Galerie de prompts IA OpenZhenZhen · MiniMax H3 · video

Vulgarisation de l'IA dans un vrai bureau

AI Knowledge Talk in a Real Study

Vulgarisation de l'IA dans un vrai bureau — découvrez cette création avec MiniMax H3 sur OpenZhenZhen, son prompt original et un exemple de génération à réutiliser.

Vulgarisation de l'IA dans un vrai bureau — AI video creation
Vulgarisation de l'IA dans un vrai bureau

Prompt

Prompt chinois

生成一段15秒、9:16竖屏、超写实真人口播视频,仅有一名成年女性。必须使用所提供的身份参考图作为唯一且必需的外观参考,高度保持她的五官、发型、肤色、服装和整体气质一致。不要请求或使用源视频或音频。人物坐在真实居家书房中,胸口以上中近景,固定机位,镜头几乎不运动。自然窗光从侧前方照射,背景书架轻微虚化,整体呈现真实手机或微单拍摄的知识创作者质感:轻微传感器噪点、真实皮肤纹理、自然呼吸、眨眼、微表情和轻微头部运动;不要磨皮、美颜滤镜、塑料皮肤、夸张电影光影、字幕、水印或任何画面文字。人物面对镜头,以精准中文口型同步自然说出:“很多人以为,AI回答问题是在数据库里找答案。其实不是,它更像是在根据前面的内容,不断预测接下来最可能出现的词。”表演节奏:0–3秒,放松地直视镜头,以聊天般语气开场,轻微眨眼。3–7秒,说到“其实不是”时轻轻摇头一次,并自然挑动眉毛。7–12秒,解释“根据前面的内容”时,一只手进入画面做小幅克制的解释手势。12–15秒,说完“最可能出现的词”后轻微点头收尾。整体表达自然即兴,不要背稿感或播音腔。眼神可以短暂偏离镜头后自然看回,手势保持克制且不频繁。

Prompt anglais

Create a 15-second, 9:16 vertical, photorealistic talking-head video featuring one adult woman. Use the supplied identity reference image as the required and exclusive appearance reference; preserve her facial features, hairstyle, skin tone, clothing, and overall demeanor. Do not request or use source video or audio. Place her seated in a real home study, framed chest-up in a medium close shot from a fixed camera with virtually no camera movement. Natural window light comes from the side-front, with a gently blurred bookshelf behind her. The image should feel like an authentic knowledge creator recording with a real phone or mirrorless camera: subtle sensor noise, realistic skin texture, natural breathing, blinking, micro-expressions, and slight head movement; no skin smoothing, beauty filter, plastic appearance, exaggerated cinematic lighting, subtitles, watermark, or on-screen text. She faces the camera and naturally speaks this Chinese dialogue with precise lip synchronization: “很多人以为,AI回答问题是在数据库里找答案。其实不是,它更像是在根据前面的内容,不断预测接下来最可能出现的词。” Performance timing: 0–3 seconds, begin with relaxed direct eye contact and conversational delivery, with a subtle blink. 3–7 seconds, on “其实不是”, gently shake her head once and make a natural eyebrow movement. 7–12 seconds, while explaining “根据前面的内容”, let one hand enter the frame for a small restrained explanatory gesture. 12–15 seconds, after saying “最可能出现的词”, give a slight nod to finish. Keep the delivery spontaneous and conversational, without a memorized feeling or announcer voice. Her gaze may briefly drift off-camera and naturally return; gestures remain restrained and infrequent.

Thèmes et styles

Voir le modèle API

Découvrir plus de prompts image et vidéo · Générer des images · Générer des vidéos · Créer sur le canevas IA