story-video-generator

To create a narrated story video, provide a short prompt and the system expands it into a structured short story with 3-8 scenes, generating scene-specific images, TTS narration, and synchronized subtitles automatically.

Automated scene generation works by expanding a user input into structured story scenes, generating scene-specific prompts for image generation, and then synthesizing TTS narration to produce a cohesive visual and audio narrative.

Yes, you need the z-ai-web-dev-sdk dependency, as the workflow relies on it to handle image generation and TTS narration required for assembling the final video files with synchronized audio.

Yes, you can add background music to TTS narration; the system blends TTS audio per scene with optional background music during the final video assembly process to create a cohesive output.

A key limitation is that story generation is constrained to 3-8 scenes per video, and the system requires specific included scripts for image generation, TTS, and video assembly to function correctly.