subject_definitions <Subject 1> (Hoshino Uyu): character… | Deepfake AI视频

作者头像
@zaizen

生成此视频所用的文字转视频提示词

subject_definitions <Subject 1> (Hoshino Uyu): character identity from user-provided image; includes pink fur, large yellow ears, tan skin, sleeveless pink tank top, dark trousers, and brown shoes. summary Task Type: reference generation. Using the provided image of "Hoshino Uyu" as a character source and an audio reference to generate an 8-second video where the subject transitions from a dim dungeon setting to a vibrant, neon cyber space while performing magic. retention_analysis <Subject 1> identity/appearance: fully_preserved across all shots. Audio (Japanese dialogue): reference for voice quality and speech timing. detailed_description [Shot 1] (0-2s): Close up of Subject 1's face. Subject 1 walking in a dimly lit, ancient stone dungeon with mossy walls and deep shadows, evoking the atmosphere of "Frieren". He is looking toward the camera, preparing his magical energy.mischievous smile while saying "<d lang='ja'>こんな場所に、こんな世界はどうかな?</d>" [Shot 2] (2-4s): A fast transition/dissolve where the dark, heavy environment of the dungeon rapidly melts away into a bright, glowing pink neon world with digital grid lines and high-tech aesthetics. [Shot 3] (4-6s): Medium shot in the vibrant pink cyber space. Subject 1 raises his hands gracefully, summoning glowing, holographic numbers "0" and "1" that swirl around him like magical particles. [Shot 4] (6-8s): Close up of Subject 1's face. He breaks into a wide,"<d lang='ja'>ふふっ...</d>" overall_soundscape The sound begins with low-frequency dungeon ambience, such as water dripping and distant stone echoes, which then transitions into high-pitched digital hums and data processing sounds during the scene shift. non_diegetic_music N/A