Three-layer audio design methodology for video storyboard pipelines. Every shot gets synchronized score, SFX, and voice — anchored to real film scores, physical parameters, and emotion tension values.

When to Use

Designing audio for storyboard shots (any genre/culture)
Writing audio field descriptions for AI video generation prompts
Choosing reference film scores for a scene's emotional tone
Parameterizing voice/dialogue delivery for TTS or voice direction
Fixing audio-visual desync in generated video

Core Rule: One Composer Per Shot

Select exactly ONE reference work per shot. Match via: scene emotion tension (0-10) → tension dimension in references/emotion-tension-map.md → pick ONE composer whose sound DNA fits the cultural context.

Do NOT stack 2-3 composers. One precise reference > three vague ones.

Three-Layer Template

When to Use

Designing audio for storyboard shots (any genre/culture)
Writing audio field descriptions for AI video generation prompts
Choosing reference film scores for a scene's emotional tone
Parameterizing voice/dialogue delivery for TTS or voice direction
Fixing audio-visual desync in generated video

Core Rule: One Composer Per Shot

Do NOT stack 2-3 composers. One precise reference > three vague ones.

Audio Score Design

When to Use

Core Rule: One Composer Per Shot

Three-Layer Template

Audio Score Design

When to Use

Core Rule: One Composer Per Shot

Three-Layer Template

Prose

Golang Patterns

Audiocraft Audio Generation

Pokemon Player

Ideation

Storybook Upgrade