dialogue-voice
Create a reusable voice bible and exact dialogue cue list from screenplay lines, including character voice identity, delivery intent, language, timing, and continuity across scenes.
SKILL.mdskills/dialogue-voice/SKILL.md
GitHubDialogue Voice
Workflow
- Read core contract,
CHARACTER_PROFILE.md, style rules, screenplay,02_screenplay/line_manifest.jsonlwhen present, character bible, canon, and continuity. - Assign one
VOICE-###identity per recurring voice. Use the canonical speech signature for cadence, habitual delivery behavior, and language style, while keeping acoustic timbre/accent/reference-audio identity in the voice bible. - Write
04_generation/voice_bible.mdwith vocal age range, register, texture, cadence, accent or dialect only when decided, speech habits, emotional range, and forbidden drift. - Write
04_generation/voice_cues.jsonlwith the originatingLINE-###, exact screenplay text, scene, speaker, language, delivery intent, timing target if known, measured duration when available, and target model. - Do not rewrite dialogue merely to make TTS easier without updating the screenplay.
Done
Every spoken LINE-### in scope maps to one voice identity and exact text source, and measured speech duration is available to the shooting script when generated.