video-storytelling

AI-generated video creation coordinates image generation with character consistency locks, synthesized narration mapped to specific voices, and automated assembly into MP4 format. This Skill generates multi-scene videos by enforcing visual style and character persistence across all scenes while synchronizing ElevenLabs voice narration to create coherent stories.

Yes. This Skill pairs OpenAI image generation with ElevenLabs text-to-speech, applying global style locks and character persistence across scenes, then assembles synchronized audio and visuals into a final MP4 using ffmpeg automation.

Visual consistency is enforced through STYLE_LOCK and NEGATIVE_LOCK parameters that persist across all scenes, plus CHARACTER_LOCK to maintain character appearance. These constraints are applied globally during image generation to produce coherent multi-scene narratives without manual retouching.

This Skill automates ffmpeg-based video assembly by accepting generated images and synthesized narration, then encoding them into standardized output (1080x1080, 30 fps, H.264 video, AAC audio) as a single MP4 file without manual encoding steps.

Yes. The Skill defaults to a title scene plus five story scenes with enforced character and style persistence, making it suited for educational content, children's stories, and social media storytelling where narrative coherence and voice mapping are critical.

Dependencies include OpenAI for image generation, ElevenLabs for voice synthesis, Pillow for image processing, and Requests for API calls. Output is standardized at 1080x1080 resolution, 30 fps, H.264 video codec, and AAC audio format in MP4 containers.