10秒動画
2026/8/9使用したAI MiniMax H3
年齢制限 全年齢
声が Kawaii かもです
いわゆるマスピ声ですが

アップスケールが雑なのでノイズが多くてすみません💦
プロンプト
subject_definitions: <Picture 1> ([Shot 1] first frame): A young woman with long red hair and a red dress stands by a window, leaning against the window frame. <Subject 1> (S1) is the female character in the image, depicted as a young woman with long red hair, wearing a red dress, and standing near a window. summary: [video continuation] + [audio reuse] + [keyframe completion] with <Picture 1> as the starting frame and the user's spoken lines as the final audio cue. retention_analysis: <Picture 1> ([Shot 1] first frame): fully_preserved - the character's appearance and pose are fully visible and must be preserved. <Subject 1> (S1) (appears in [Shot 1]): fully_preserved - the character's visual identity and pose are fully preserved. <Audio 1> (appears in [Shot 1]): weak_reference - the audio is not referenced in the visual frame. detailed_description: The video begins with the young woman with long red hair and a red dress standing by the window, leaning against the window frame. She looks directly at the camera with a calm expression. Then, she raises her hand in a wave and immediately begins to spin 360 degrees on the spot, moving away from the camera. As she spins, the camera follows her movement with a smooth tracking shot, capturing her as she runs down the hallway. She continues running until she reaches the end of the hallway, then sharply turns around 180 degrees and runs back toward the camera. As she approaches, she stops just before reaching the camera and raises her hand to make a peace sign. She then says: <d>[Japanese] ぴーす!ぴーす!!</d> The camera remains static as she makes the peace sign, and the scene ends with her expression and gesture clearly visible. overall_soundscape: The video features no ambient sound or physical action noise; the only sound is the character's voice, which is clearly audible and directly tied to her actions. The scene is silent except for the character's voice, which is the only diegetic sound. non_diegetic_music: A light, upbeat electronic track with a fast tempo and a driving rhythm, designed to match the energetic movement of the character. The music is brief and ends as the character makes the peace sign.
ネガティブプロンプト
Steps
Scale
Seed
Sampler
Noise
Strength
コメント

 
おかしら おかしら 2023年3月より活動開始 271投稿46フォロー91フォロワー ローカル生成専門で遊んでいます 現在は動画のローカル生成の勉強中です (HunyuanやWanなど) ---- GPU以外も新調しました! Core i5-12400F RAM128G RTX3060VRAM12G ---- 当方いわゆるポン出し専門です いろいろお目汚しな点も多くあることをご承知おきください ---- 生成AI各種OSSやツールの製作者諸氏には感謝です