What problem does it solve? Writing prompts for generative audio models is error-prone: voices drift between segments, sound effects get ignored, and dialogue falls out of sync with the storyboard. This Skill provides a structured five-part method for authoring voice profiles and full-chapter audio scripts that the seed-audio-1.0 model can perform and mix in one pass. ## Core Features & Use Cases - Voice Profile Design: Write six-slot voice descriptions (age, register, timbre, pace, accent, temperament) derived from character sheets for custom voice casting. - Chapter Audio Scripts: Draft segment-by-segment scripts with per-line delivery directions, music descriptions, inline sound effects, and relative timestamp control aligned to shot durations. - Voice Anchoring: Bind characters to reference audio so the same voice stays consistent across an entire production. - Use Case: Given a chapter's shot list, run the draft command to generate a baseline script, then refine tone, breathing details, and music cues per line before generating a fully mixed audio track. ## Quick Start Ask the assistant to draft an audio script for a chapter using the score command, then refine the voice descriptions and per-line delivery notes before generating the mixed track.
