Music and Voices

Status: Candidate production standard

The working music pipeline is strongest when a song is designed for two distinct voices, uses explicit spoken-word tags, and describes the emotional movement of each section.

Voice functions

Voice	Dramatic job
Lumi	Ground the idea, model listening, carry reassurance and wonder
Pip	Create momentum, ask the active question, embody surprise and play
Together	Deliver the hook and the shared discovery

The functions can swap for story reasons, but the voices should never feel interchangeable.

Song brief structure

song:
  episode: S01E__
  letter: ""
  duration_target: ""
  emotional_arc: curious -> uncertain -> delighted -> warm
  spoken_word:
    opening: "[Lumi, softly spoken]"
    pivot: "[Pip, excited spoken interjection]"
  sections:
    - intro
    - call_and_response_verse
    - chorus
    - discovery_bridge
    - final_chorus
    - gentle_button
  hook: ""
  anchor_words: []
  generation_record:
    tool: Suno
    prompt_version: ""
    output_link: ""
    status: Draft

Rules

• Give each voice an intention and feeling, not just a name.
• Use spoken tags where story clarity matters.
• Let the chorus express the discovery, not merely repeat the letter.
• Write visual beats and song structure together.
• Preserve generation prompts and selected output links.